Introduction to Computer Science in C#

Release 2.0

Andrew N. Harrington and George K. Thiruvathukal

15-July-2013 15:10:48
 CONTENTS

1 Context 1
1.1 Motivation for This Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Resources Online . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Downloading Text and Source Code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.4 Introduction in Miles Chapter 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2 C# Data and Operations 5

2.1 A Sample C# Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2 Lab: Editing, Compiling, and Running with Xamarin Studio . . . . . . . . . . . . . . . . . . . . . . 8
2.3 Arithmetic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.4 Variables and Assignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.5 Syntax Template Typography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.6 Strings, Part I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.7 Writing to the Console . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.8 C# Program Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.9 Combining Input and Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.10 String Special Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
2.11 Substitutions in Console.WriteLine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.12 Value Types and Conversions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.13 Learning to Solve Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.14 Lab: Division Sentences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

3 Defining Functions of your Own 41

3.1 A First Function Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.2 Multiple Function Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.3 Function Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.4 Multiple Function Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.5 Returned Function Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.6 Two Roles: Writer and Consumer of Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.7 Local Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.8 Static Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.9 Not using Return Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.10 Library Classes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
3.11 Static Function Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

4 Basic String Operations 55

4.1 String Indexing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4.2 Some Instance Methods and the Length Property . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4.3 A Creative Problem Solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
4.4 Lab: String Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

i
5 Decisions 61
5.1 Conditions I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
5.2 Simple if Statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
5.3 if-else Statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
5.4 More Conditional Expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
5.5 Multiple Tests and if-else Statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
5.6 If-statement Pitfalls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
5.7 Compound Boolean Expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

6 While Loops 73
6.1 While-Statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
6.2 While-Statements with Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
6.3 Interactive while Loops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
6.4 Short-Circuiting && and || . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
6.5 While Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
6.6 More String Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
6.7 User Input: UI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
6.8 Greatest Common Divisor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
6.9 Do-While Loops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
6.10 Number Guessing Game Lab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

7 Foreach Loops 101

7.1 foreach Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
7.2 foreach Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102

8 For Loops 103

8.1 For-Statement Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
8.2 Examples With for Statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
8.3 Lab: Loops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

9 Files, Paths, and Directories 117

9.1 Files As Streams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
9.2 Writing Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
9.3 Reading Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
9.4 Path Strings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
9.5 Directory Class . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
9.6 File Class . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
9.7 Command Line Execution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
9.8 FIO Helper Class . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124

10 Arrays 127
10.1 One Dimensional Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
10.2 Musical Scales and Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
10.3 Linear Searching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
10.4 Sorting Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
10.5 Binary Searching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
10.6 Lab: Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
10.7 Lab: Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
10.8 Multi-dimensional Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150

11 Lists 155
11.1 List Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
11.2 .Net Library (API) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158

12 Dictionaries 159

ii
 12.1 Dictionary Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
12.2 Dictionary Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
12.3 Dictionary Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
12.4 Lab: File Data and Collections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

13 Classes and Object-Oriented Programming 165

13.1 A First Example of Class Instances: Rational . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
13.2 Classes And Structs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
13.3 Class Instance Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175

14 Testing 181
14.1 Assertions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
14.2 Attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
14.3 Testing the Constructor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
14.4 Testing Rational Comparisons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
14.5 Testing Rational Arithmetic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
14.6 Testing Rational Conversions (to other types) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
14.7 Testing the Parsing Feature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
14.8 Running the Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184

15 Interfaces 185
15.1 Rationals Revisited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
15.2 csproject Revisited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
15.3 Group Game Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188

16 Recursion 195

17 Data Structures 197

18 Appendix 199
18.1 Development Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
18.2 Xamarin Studio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
18.3 Command Line Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
18.4 Precedence of Operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
18.5 Homework: Grade Calculation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
18.6 Homework: Grade Calculation from Individual Scores . . . . . . . . . . . . . . . . . . . . . . . . . 211
18.7 Homework: Grade File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
18.8 Homework: Book List . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
18.9 Lab: Version Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
18.10 Mercurial and Teamwork . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
18.11 Recent/Current Course Offerings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
18.12 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235

Bibliography 237

Index 239

iii
iv
 CHAPTER

ONE

CONTEXT

1.1 Motivation for This Book

This is really a preface, but the otherwise very capable Sphinx publishing environment that we use is not set up for a
separate preface.

1.1.1 Pedagogy

Our first aim is to provide a good introduction and conceptual framework for more computer science, not an ency-
clopedic coverage of C#. C# will not be most students’ only language, or necessarily the most used. Designing and
creating algorithms in a particular language is an important skill, requiring ongoing effort, so most of the text is still
centered on C#.
C# is an object-oriented language. There is the ongoing argument about when to introduce details of object-oriented
programming. We last taught Java, objects first. Students dutifully followed our lead. Later, we saw quick programs
that students wanted to write for themselves, that were layered with totally unnecessary and distracting instances.
We have seen less problem with the claimed opposite issue: starting with basically procedural programming, and then
some time after starting OOP, still not seeing where design with objects is useful.
In any event, we start off with more procedural programming, then introduce the use of instances of existing classes
of objects, and then move to designing classes with instance variables, constructors, and instance methods.
We tend to introduce examples first, and then the general syntax, and then more examples and exercises.

1.1.2 C# and Mono

We have taught introductory programming for many years, through a progression of programming languages. Our last
language was Java, still the language of the AP test, which drives so many introductory texts.
We had C# in mind: It is a more modern language. Its designers got to reflect on the glitches with Java, and address
them effectively.
The key problem with C# used to be that it was totally a Microsoft language for Windows. Many of our students have
their own machines: many are OS-X machines from Apple; some are Linux. We did not want to cut those students
off. Nor did we want to limit students to thinking of a computer as a Windows machine. Meanwhile the open source
implementation of C#, Mono, has been maturing, along with its tool chains.
While many open-source tools have hackers jumping in to eliminate bugs, and maybe providing enough documen-
tation for a professional, documentation for a beginner is often lacking. This book contributes there, partly in the
documentation for Mono’s lovely interactive environment csharp, and also for the integrated development environ-
ment, Xamarin Studio. We show beginners how to start using the Xamarin Studio environment, with its large array of
features (not all needed by the beginner), and introduce more features as needed.

1
Introduction to Computer Science in C#, Release 2.0

We aim to end up with a book that provides a solid conceptual framework for beginning computer scientists in the
context of the clean, well-established modern language, C#, using multi-platform free and open-source tools, with
clear documentation.
We hope that you find this to be a winning combination.

1.2 Resources Online

This book is designed for Comp 170 at Loyola University, Chicago. The materials are available to all on the web. Here
are some important web links:
• The course example file, source.zip is the essential resource to download and unzip onto your machine. Com-
puter programs are designed to run on a computer and solve problems. Though the initial problems will be tiny
and often silly, they will serve as learning tools to prepare for substantive problems.
• http://introcs.cs.luc.edu is an online text version for your web browser. See also Downloading Text and Source
Code. Except on very geometrically oriented topics, text-oriented learners may be happiest just reading the
book.
• https://luc.box.com/s/ggx0n1a0ijf8ctikxjo4/1/611829684 is a box.com folder containing all the videos. The
numbers at the beginning of the titles are chapter and section numbers from the text. The listing in box.com
takes up several pages. These box.com videos may be
– streamed, including at full size (though generally after an initial delay),
– downloaded individually or
– the whole gigabyte folder can be downloaded at once to play later on your machine. (This is a choice on
the menu under the page’s Folder Options.)
There is a mixture of new high-def videos and older lower resolution 800x600 pixel videos. See
00README.html in that folder for a description of the differences between the old videos and the latest update
of the online book.
Box does not let you stream videos automatically in a playlist sequence. Most of the
videos for the beginning of the book are created more recently in high-definition, in
a form that YouTube handles in both full or lower resolution in the YouTube playlist
https://www.youtube.com/playlist?list=PLc5TVkj91XZhxRSJHsDoJzX5K1XKBAFLH. The older videos’
size does not play well on YouTube, so only the newer videos are in the playlist. This is a good place to play
through at least the first five chapters of the book, plus some isolated later sections’ videos. YouTube does not
supply a direct way to download and save videos, so use the Box folder for that.
For those who learn best with spoken words combined with written words, the videos should be a good start.
Even if you use a video for a section, you are encouraged to review the written text afterward. Then be aware of
the written version for quick reference. The written text may include extra details and exercises, and it will have
the latest revisions.
In various formats, be aware of these helpful features:
• We have picked out particularly important words, phrases, and symbols, and put them in our index, which is
accessible from each web page.
• In a web version on our website, you can use the Search option to look for words, in general. The location of
the search link varies depending on theme being used, and can be hidden. This does not work on a downloaded
local html copy.
• We start with a brief table of contents for the whole book. In the web versions you can get the most detailed
table of contents for a single chapter by clicking on a chapter title in the main table of contents or the current
chapter title in the top banner on a regular content page.

2 Chapter 1. Context
 Introduction to Computer Science in C#, Release 2.0

Her are further links that may be useful in our repository:

• http://bitbucket.org/loyolachicagocs_books/introcs-csharp/ is the home page for the repository of all the sources
for the book. To read this book, you do not need to go to that URL, but if you do, the home page gives you an
idea of what updates have been made recently to the book or accompanying examples. Since improvements are
made on an ongoing basis, the notes about recent changes may be useful to you.
The multiple production versions are generated largely by Sphinx software from the common set of sources in
the repository. The sources are largely plain text files.
• http://bitbucket.org/loyolachicagocs_books/introcs-csharp/src is the part of our repository containing the latest
versions of the source code files. You can quickly browse and view individual files in the examples folder.
Example file links throughout this text refer to these repository files.

1.3 Downloading Text and Source Code

1.3.1 Primary Site

The following table provides links for alternate web viewing, printing, or e-reading. These are provided at the master
site, http://introcs.cs.luc.edu.

Table 1.1: Course Text Reading/Download Options

Format URL
Web Pages/Default Theme http://introcs.cs.luc.edu/book/latest/html/default/
Web Pages/Responsive Theme http://introcs.cs.luc.edu/book/latest/html/bootstrap
Zip archive http://introcs.cs.luc.edu/book/latest/download/default/html.zip
PDF http://introcs.cs.luc.edu/book/latest/download/default/comp170.pdf
ePub Experimental http://introcs.cs.luc.edu/book/latest/download/default/comp170.epub
Table 1.2: Source Code Reading/Download Options
Format URL
C# Examples (as pages) http://bitbucket.org/loyolachicagocs_books/introcs-csharp/
C# Examples (as ZIP) https://bitbucket.org/loyolachicagocs_books/introcs-csharp-examples/get/default.zip

1.3.2 Dr. Thiruvathukal’s Site

Dr. Thiruvathukal maintains an offsite web presence for his own site, http://thiruvathukal.com, which is a reliable
mirror to the Primary Site above.

Table 1.3: Available Formats (Primary Site)

Format URL
Web Pages http://introcs.courses.thiruvathukal.com/html/
Web Pages (offline ZIP) http://introcs.courses.thiruvathukal.com/dist/html.zip
PDF http://introcs.courses.thiruvathukal.com/latex/comp170.pdf
ePub (Experimental) http://introcs.courses.thiruvathukal.com/epub/comp170.epub
Table 1.4: Source Code Reading/Download Options
Format URL
C# Examples (as pages) http://bitbucket.org/loyolachicagocs_books/introcs-csharp/
C# Examples (as ZIP) https://bitbucket.org/loyolachicagocs_books/introcs-csharp-examples/get/default.zip

1.3. Downloading Text and Source Code 3

Introduction to Computer Science in C#, Release 2.0

1.4 Introduction in Miles Chapter 1

For a general introduction to programming and the context of C# in particular, there is already a good free online
source, Rob Miles’ C# Yellow Book. To download it, go to http://www.csharpcourse.com/ and click on the link for
Rob Miles CSharp Yellow Book 2012.pdf (or the latest year).
Read Miles Chapter 1, being aware of the comment below:
The chapter makes some reference to Microsoft, the original creator of C#, and its Visual Studio software development
environment, which works only on Windows machines, and costs a lot if you are not a student. The Lab: Editing,
Compiling, and Running with Xamarin Studio will introduce an alternative to the Microsoft environment: Xamarin
Studio and Mono, which are free, open-source software projects that make C# available for multiple platforms: Win-
dows, Mac, or Linux machines. With a substantial fraction of students having their own machine that does not run
Windows, this flexibility is important.

4 Chapter 1. Context
 CHAPTER

TWO

C# DATA AND OPERATIONS

2.1 A Sample C# Program

In the examples that you should have downloaded is a first simple program, painting/painting.cs.
Here is what it looks like when it runs, with the user typing the 20.5 and the 10:
Calculation of Room Paint Requirements
Enter room length: 20.5
Enter room width: 10
The wall area is 488 square feet.
The ceiling area is 205 square feet.

This is not very exciting, but it is a simple place to start seeing basic program features. We will refer back to this
sample run while discussing the program. Here is the text of the program:
This section gives an overview of a working program, even if all the explanations do not make total sense yet. This is
a first introduction of concepts and syntax that gets fully explained in further sections.
Do not worry if you not totally understand the explanations! Try to get the gist now and the details later.
The different colors are used in modern program editors to emphasize the different uses of the parts of the program.
We give a line by line explanation:
The C# environment supplies an enormous number of parts that you can reference. Nobody is familiar with all of
them. If you had to make sure you always used names that did not conflict with other names supplied, you would be
in trouble. To avoid this C# has namespaces. The same name can be used in different namespaces without conflict.
The central standard namespace is System. We will always include this first line, using System;.
Lines 2, 10, 18, and 21 are blank. This is merely for the human reader to separate sections visually. The computer
ignores them.
A basic unit in C# is a class. Our code sits inside a class. Each class has a heading with class followed by a name.
This class is Painting. After the heading comes a body delimited by braces. The opening brace { in line 4, is
matched by the closing brace } on the last line of the program.
A class is broken up with chunks called functions or methods. Each has a heading. C# allows the currently popular
programming paradigm called object-oriented programming, where classes generally describe new kinds of objects.
This is useful in complicated situations, but we start more simply with the older procedural programming. Unfortu-
nately for now, the more common situation is with objects, so a function that does not involve such new objects must
be marked specially as static.
Functions can be like in math, where they produce a function value for later use. In C# they can also just do something
(like write to the screen), and not produce a value for later use in the program. To show that no function value is
produced, the word void is used.

5
Introduction to Computer Science in C#, Release 2.0

Every program must start running somewhere. In C# that is at a function with name Main. So our program starts
running here. This syntax for this function needs to start just like here, with static void Main.
Even though this is not a mathematical function producing a value, a function in C# must be followed by parentheses
( ).
After the function heading comes a body. Like with a class, a function body is delimited by braces. The opening brace
here is matched by the closing brace on the second to last line of the program.
A program works with data of many different possible types. One type is double. A double can hold a number,
including a possible fractional part.
To refer to data in a program we use names called variables. This line says that width, length, wallArea, and
ceilingArea are all the names for variables that can hold a double value. We will assign values to these variables
later.
This line is a declaration statement. Most statements in C#, like this one, end with ; - a semicolon.
This is another declaration. This time the type of the variables is string, which means a sequence of characters, like
a line you might type at the keyboard.
Here is another declaration for a double, looking slightly different. In this case we follow a convention, using all
capital letters, to suggest that the value of HEIGHT will be constant (unchanging), and we assign its value at the same
time with = 8. This naming of constants is not strictly necessary, but it makes the program easier to read.
Console refers to the terminal or console window where text output appears for the program. One of the things you
can do with the Console is WriteLine, to write a line. The period between Console and WriteLine indicates
WriteLine is a named part of the Console. This WriteLine is a function. Like in math, it can have a parameter
in parentheses. While you are used to a parameter for a function in math being a number, functions in C# are much
more general. A function can be defined with any type of parameters. Here the parameter is a string, "Calculation
of Room Paint Requirements", delimited by the quotes at either end. Notice that the contents of this string
appear at the start of the screen output displayed for this program. The program did write this line.
This statement is similar to the last one, except that it uses Write rather than WriteLine. The WriteLine
function wrote a whole line - see that the output next after the WriteLine statement started on the next line. Here
Write does not advance the printing position to the next line after it: The 20.5 of the sample out follows the string on
the same line.
This statement serves as a prompt: letting the user know that information is being requested (a room length).
Here is where the program takes in the information requested from the user. Its action is actually right to left:
Console.ReadLine is another function available with the Console, that reads a line typed in by the user on
the keyboard. Here in the sample run, on the same line as the prompt string, the user types 20.5 and the Enter or
Return key.
In the sample run, the value produced by the Console.ReadLine function is these four characters 20.5.
Recall that lengthString was declared as a variable to hold a string. The = indicates an assignment statement.
It is an assignment of the value on the right of the equal sign to be the current value of the variable on the left of
the equal sign. In the sample run, this would mean that the variable lengthString would end up holding the
value "20.5". Though these characters happen to look like a number, any sequence of characters can be typed. The
Console.ReadLine() function produces this sequence of characters as a string type.
Of course we want to interpret the user’s input as a number in order to do our arithmetic. This line makes the conversion
between the types.
It is another assignment statement (with the =). We are assigning to the variable length, which we declared as a
double. The value assigned comes from the expression on the right of the =, double.Parse(lengthString).
The function double.Parse, is just the one we want, it takes a string parameter lengthString containing the
string from the user input, and the value produced is the corresponding double number. In the sample run that
assigns to length the value 20.5.

6 Chapter 2. C# Data and Operations

 Introduction to Computer Science in C#, Release 2.0

These lines are analogous to the previous three lines: give a prompt for the user; get the user response; convert it to a
double, and assign to a variable (width in this case). In the sample run the variable width is assigned the value
10.
At this point we have all the data we need from the user. The next part is the brief calculations of results:
At the end of the first line is a comment. It starts with // and ends at the end of the same line. It is ignored by the
compiler. It is there for humans, hopefully to add something that helps understanding of the program.
We have two assignment statements. The values to assign are given by arithmetic expressions on the right side of the
equal signs. It looks pretty much like regular math, except in math class you may be used to only having one letter
names for variables, unlike length, width, and HEIGHT.
The tradeoff for allowing multiple character names is that multiplication must have an explicit operation symbol. The
symbol used for multiplication in C# is * an asterisk. The + and parentheses serve their normal mathematical purpose.
In the sample run, the value of 2 * (length + width) * HEIGHT is
2 * (20.5 + 10) * 8
which simplifies to 488.
With the sample run, ceilingArea would get the value 20.5 * 10, or 205.
This is a single statement. Line endings act just like a space in C#. The statement ends with the semicolon on the
second line.
Again Console.WriteLine will print something to the computer console. This time the string printed is more
complicated: It starts off with the literal string "The wall area is ", but then we want to print out the calculated
result. The + wallArea allows that. The + sign after the string is not a mathematical operator here. Coming after
a string, it has a special string meaning: It converts the next part wallArea to be a string. In the sample run that
would be converting the double value 488 to be the string "488". The plus sign then “adds” the strings in a manner
appropriate for strings, concatenating them. That means joining them together, end to end.
The + " square feet." then tacks on the last string. In the sample output you see what is printed:
The wall area is 488 square feet.

sandwiching the value taken from the variable wallArea between two literal string, given in quotes.
This statement behave like the previous one, except with different quoted strings and the value of a different variable.
See the sample output.
Finally we have the matching closing braces marking the end of the body of the Main function and the end of the
body of the Painting class.
Of course the display would look different if the user entered different data. Here is what is displayed when the user
enters length 15 and width 6.5:
Calculation of Room Paint Requirements
Enter room length: 15
Enter room width: 6.5
The wall area is 344 square feet.
The ceiling area is 97.5 square feet.

The blank space in the program was there to aid human understanding. In a C# program whitespace is any consecutive
combination of spaces, newlines, and tabs. C# treats any amount of whitespace just the same as a single space, except
inside quoted strings, where every character is important.
Also the compiler does not require whitespace around special symbols like {};().=*+,. Hence the paint-
ing/painting.cs program above would be just as well translated by the compiler if it were written as:

2.1. A Sample C# Program 7

Introduction to Computer Science in C#, Release 2.0

using System;class Painting{static void Main(){double width,length

,wallArea,ceilingArea;string widthString,lengthString;double HEIGHT
=8;Console.WriteLine("Calculation of Room Paint Requirements");Console.
Write("Enter room length: ");lengthString=Console.ReadLine();length=
double.Parse(lengthString);Console.Write("Enter room width: ");
widthString=Console.ReadLine();width=double.Parse(widthString);wallArea
=2*(length+width)*HEIGHT;ceilingArea=length*width;Console.WriteLine(
"The wall area is " + wallArea
+" square feet.");Console.WriteLine
("The ceiling area is "
+ceilingArea+
" square feet.");}}

Since human understanding is very important, we will emphasize good whitespace conventions, and expect you to use
them.
Next we give you an even simpler program to run in the lab. After that we return to how you can get the painting
program to run on your computer.

2.2 Lab: Editing, Compiling, and Running with Xamarin Studio

This first lab is aimed at taking you through the end-to-end process of writing and running a basic computer program
with the Xamarin Studio environment. As with all things in life, we will learn in this lab that becoming a programmer
requires you to learn a number of other things along the way.
In software development/engineering parlance, we typically describe a scenario as a workflow, which can be thought
of as a series of steps that are possibly repeated. The workflow of programming can loosely be defined as follows:
1. Use a text editor to write your source code (human readable).
2. Compile your code using the Software Development Kit (SDK) into object code.
3. Link your object code to create an executable. (There are other kinds of results to produce, but we will start with
the idea of an executable program to keep things simple.) The default is to have an executable program created
with compilation, automatically.
4. Run your program. Even for the most seasoned developers, your program may not work entirely right the first
time, so you may end up repeating these steps (debugging).
These steps can all be done with different tools. Many find it simpler to have an integrated tool, like Xamarin Studio,
that does them all in the same place, and automates the steps that do not need human interaction!
If you are doing this on your own machine, make sure you have Mono and Xamarin Studio installed as in Development
Tools.
Other tools are available, like the development environment Visual Studio (from Microsoft, only for Windows).
Understanding the lower level tools that accomplish each step is important, but we defer a discussion to get you going
with Xamarin Studio.

2.2.1 Goals

Our primary goal to create a C# Solution that you can use to do all of the remaining homework assignments and labs
this semester. If you wish, you can create as many solutions as you like, but C# allows you to create a single solution
and add (at any time) projects to it. This will provide by far the best experience for you in the course, where you can
keep adding onto previous efforts without having to start over each time.

8 Chapter 2. C# Data and Operations

 Introduction to Computer Science in C#, Release 2.0

2.2.2 Steps

We start by creating a solution with a project in it. The images are from a Mac. Windows versions should be similar.
1. Open Xamarin Studio, in the appropriate way for an application in your operating system. It should be in the
Start menu for Windows. Using Spotlight is quick on a Mac.
2. You get a Welcome screen. Toward the upper left corner is a link for New Solution. Click on it. Alternately you
can follow the path through the menus Go File -> New -> Solution.

3. You get a dialog window to fill out. Follow the order below. Later parts may not be visible until you do the
previous parts:
• Select C# in left hand side panel
• Select Console Project in the middle panel
• In the bottom field, “Solution name” (not the top Name field), enter any name you like: We recommend
work, which will make sense for all your work for the course.
• Leave the Location field above it as is or change it if you like.
• Above that, Enter hello in the Name field, for the name of the project.
• Make sure Create directory for solution is checked in the bottom right.
• Press the OK button.

2.2. Lab: Editing, Compiling, and Running with Xamarin Studio 9

Introduction to Computer Science in C#, Release 2.0

You now have created a solution in Xamarin Studio, with one project inside it. Later we can add further projects
to this solution.
4. Look at the Xamarin Studio window that appears. It should have two main sub-windows or “Pads” as Xamarin
Studio calls them. A narrow one on the left is the Solution Pad, containing a hierarchical view of the solution.
You should see your solution name at the top and the hello project under that. Folders have a little triangle
shown to their left. You can click on the triangle. A triangle pointing down means the inside of the folder is
displayed. A triangle pointing to the right means the contents are not being displayed. Listed under hello are
References and Properties, that we will ignore for now. Below them is the line for the automatically generated
sample code file Program.cs. The file should also appear in the Edit Pad to the right.

10 Chapter 2. C# Data and Operations

 Introduction to Computer Science in C#, Release 2.0

5. Program.cs should be selected in the Solution Pad, as shown above. Change the selection by clicking on hello.
At the right end of the highlighted hello entry you should see an icon with a small gear and a triangle. Click on
it to get the context sensitive popup window. When selected, most entries in the Solution Pad should show this
icon, allowing you to open its context sensitive menu.
6. Bring up the context menu on the hello project in the Solution Pad. Select Run Item.

7. Here Xamarin Studio combines several steps: saving the file, compiling it into an executable program, and
starting running it if compilation succeeded. With the canned file it should succeed! You see a Console window
something like

You have a chance to see the output of this simple program. Follow the instructions and press the space or Enter
key.
8. On Windows, that kills the window. On a Mac, only, there is one more step:

You have to actively close the Mac terminal window, either by clicking the red window closing button, or using
the keyboard, with Command-W.
9. Initially, for immediate practice running a program, this automatically generated file, Program.cs, is conve-
nient. Hereafter it is an annoyance. The file name is always the same, and not useful, and you would need to
redo the whole code for your own program. A general approach is to delete this file and put in a file of your
own:

2.2. Lab: Editing, Compiling, and Running with Xamarin Studio 11

Introduction to Computer Science in C#, Release 2.0

• Make sure Program.cs is selected in the Solution Pad. You save a step by closing the Edit Pad for Pro-
gram.cs, clicking on the X in the Program.cs tab at the top of the Edit Pad.
• In the Solution Pad open the context sensitive menu for Program.cs, and select Remove.

• You get another popup. When it appears the right button is selected, but you do not want that selection,
Remove From Project. The image below shows the proper button, the left button*, Delete, being chosen.
Otherwise the file is left in the hello folder, but it is just not listed as being in the project.

• If you forgot to close the Edit Pad tab containing Program.cs earlier, you can still do it, just say not to save
changes to the file when asked.
10. To get in code that you want, there are several approaches. The one we take now is to start from a completely
new empty file: Pop up the context sensitive menu for the hello project. Select the submenu Add... and then
New File....

12 Chapter 2. C# Data and Operations

 Introduction to Computer Science in C#, Release 2.0

11. In the popup New File Dialog Window, click on Empty File (not Empty Class). Enter the name hello.cs. Click
the New button.

12. This should add hello.cs to the hello project and open an editing window for hello.cs. The file should have no
text.

2.2. Lab: Editing, Compiling, and Running with Xamarin Studio 13

Introduction to Computer Science in C#, Release 2.0

Much like in most word processors type in (or paste) the following code. This is actually an equivalent Hello,
World! program to the automatically generated one, but it is a bit shorter. It only introduces the syntax we
actually need at the beginning, and will be discussing more shortly:
This program is deliberately simple, so you can type it into the text editor quickly and become familiar with
how to create, edit, and save a program.

13. You can run the project just as before. You should ge the same result, unless you made a typing error. In that
case look, fix it, and try again.
14. Now try a bit of editing: Look at the program to see where output came from. Change what is printed and run
it, but don’t eliminate the console window (so you can show it off).
15. Now grab the instructor or teaching assistant so they can perform a quick inspection of your work and check it
off (including the varied message printed).
Labs need to be completed to receive credit. If you are unable to make class on a lab day, please make sure that you
complete the work and demonstrate it by the beginning of the next lab.
At this point, you have accomplished the major objective for this introductory lab: to create a Xamarin Studio project,
and enter, compile, and run a C# program.

For further reinforcement

1. Can you make a new program variant print out two separate lines?
2. Download and install Mono Software Development Kit and Xamarin Studio on your home computer or laptop.

14 Chapter 2. C# Data and Operations

 Introduction to Computer Science in C#, Release 2.0

3. You can now add further projects to your current solution. To add a new project in your solution, in the Solution
Pad open the context sensitive menu for the whole solution (top line), select Add, and in the submenu select
New project.
You see a window much like when creating a solution, except there is no line for a solution name. Complete the
remaining parts in the same way, giving a new name for the project.

2.3 Arithmetic

We start with the integers and integer arithmetic, not because arithmetic is exciting, but because the symbolism should
be mostly familiar. Of course arithmetic is important in many cases, but C# is often used to manipulate text and other
sorts of data.

2.3.1 Csharp

Of course we could write programs to demonstrate arithmetic, but there is a fair amount of overhead with a full
program. For just testing little bits, there is another alternative: The Mono system comes with a program csharp. Let
us try it out.
Open a terminal (Linux or OS X) or Mono Command Prompt window in Windows, and enter the command csharp.
You should see :
Mono C# Shell, type "help;" for help

Enter statements below.

csharp>

The csharp> prompt tells you that the C# interpreter has started and is awaiting input. This allows you to create
small bits of C# and test them, interactively, without having to write a full program!
Play along with the examples here, entering what comes after the prompt:
csharp> 2 + 3;
5

The csharp program just has a read, evaluate, and print loop: the acronym is repl. It evaluated the expression 2 + 3
and printed the result, on a line without a prompt. Csharp can evaluate arbitrary C# expressions. It is very handy for
testing as you get used to new syntax.
Subtraction works as you would expect. Blanks are optional around symbols:
csharp> 10 - 3;
7

For the binary arithmetic operators, you are encouraged to add blanks to make the expression more easily readable by
humans.
The csharp program is not line-oriented. The semicolon indicates that you are finished with an entry. You can easily
forget it. If your statement is incomplete you get another > prompt (with no “csharp”), until you complete your entry
with a ; (semicolon).
csharp> 10-3
> ;
7

In math class you could enter something like 4(10) for multiplication:

2.3. Arithmetic 15
Introduction to Computer Science in C#, Release 2.0

csharp> 4(10);
{interactive}(1,2): error CS0119: Expression denotes a ’value’,
where a ’method group’ was expected

Unfortunately the error messages are not always easy to follow: it is hard to guess the intention of the user making a
mistake.
The issue here is that the multiplication operator must be explicit in C#. Recall that an asterisk is used as a multiplica-
tion operator:
csharp> 4 * 10;
40

Enter each of the following expressions into csharp, and think what they will produce (and then check):
2*5;
2 + 3 * 4;

If you expected the last answer to be 20, think again: C# uses the normal precedence of arithmetic operations: Mul-
tiplications divisions, and negations are done before addition and subtraction, unless there are parentheses forcing the
order:
csharp> -(2+3)*4;
-20

A sequence of operations with equal precedence also work like in math: left to right.
csharp> 10 - 3 + 2;
9

2.3.2 Division and Remainders

We started with the almost direct translations from math. Division is more complicated. We continue in the csharp
program:
csharp> 5.0/2.0;
2.5
csharp> 14.0/4.0;
3.5

So far so good. Now consider:

csharp> 14/4;
3

What? Some explanation is in order. All data has a type in C#. When you write an explicit number without a decimal
point, like 2, 17, or -237, it is interpreted as the type of an integer, called int for short.
When you include a decimal point, the type is double, representing a more general real number. This is true even if
the value of the number is an integer like 5.0: the type is still double.
Addition, subtraction, and multiplication work as you would expect for double values, too:
csharp> 0.5 * (2.0 + 4.5);
3.25

Note: In C#, the result of the / operator depends on the type of the operands, not on the mathematical value of the
operands.

16 Chapter 2. C# Data and Operations

 Introduction to Computer Science in C#, Release 2.0

If one or both of the operands to / is a double, the result is a double, close to the actual quotient. We say close,
because C# stores values with only a limited precision, so in fact results are only approximate in general. For example:
csharp> 1.0/3;
0.333333333333333

Small errors are also possible with the double type and the other arithmetic operations. See Type double.
Division with int data is handled completely differently.
If you think about it, you learned several ways to do division. Eventually you learned how to do division resulting in
a decimal. In the earliest grades, however, you would say
“14 divided by 4 is 3 with a remainder of 2.”
Note the the quotient is an integer 3, that matches the C# evaluation of 14/4, so having a way to generate an integer
quotient is not actually too strange. The problem here is that the answer from grade school is in two parts, the integer
quotient 3 and the remainder 2.
C# has a separate operation symbol to generate the remainder part. There is no standard single operator character
operator for this in regular math, so C# grabs an unused symbol: % is the remainder operator. (This is the same as in
many other computer languages.)
Try in the csharp shell:
csharp> 14 / 4;
3
csharp> 14 % 4;
2

You see that with the combination of the / operator and the % operator, you get both the quotient and the remainder
from our grade school division.
Now predict and then try each of these expression in csharp:
23/5;
23%5;
20%5;
6/8;
6%8;
6.0/8;

Finding remainders will prove more useful than you might think in the future! Remember the strange % operator.

Note: The precedence of % is the same as / and *, and hence higher than addition and subtraction, + and -.

When you are done with csharp, you can enter the special statement
quit;
There are some more details about numeric types in Value Types and Conversions.
We have been testing arithmetic expressions, with the word expression used pretty much like with normal math. More
generally in C# an expression is any syntax that evaluates to a single value of some type. We will introduce many more
types and operations that can be used in expressions.

Divisible by 17 Exercise

What is a simple expression that lets you see if an int x is divisible by 17?

2.3. Arithmetic 17
Introduction to Computer Science in C#, Release 2.0

Mixed Arithmetic Exercise

Think of the result of one of these at a time; write your prediction, and then test, and write the correct answer afterward
if you were wrong. Then go on to the next.... For the ones you got wrong, can you explain the result after seeing it?
2 * 5 + 3;
2 + 5 * 3;
1.5 * 3;
7.0/2.0;
7.0/2;
7/2.0;
4.0 * 3 / 8;
4 * 3 / 8;
6 * (2.0/3);
6 * (2/3);
3 + 10 % 6;
10 % 6 + 3;

2.4 Variables and Assignment

Each piece of data in a C# program has a type. Several types have been introduced: int for integers, double for
numbers allowing a fractional part, approximating more general real numbers. There are many other numeric types and
also non-numeric types, but we can use int and double for examples now. Data gets stored in computer memory.
Specific locations in memory are associated with the type of data stored there and with a name to refer to it.
A program allocates a named storage spot for a particular type of data with a declaration statement, like:
int width;

Each declaration must specify a type for the data to be stored and give a name to refer to it. These names associated
with a data storage location are called variables.
The declaration statement above sets aside a location to store an int, and names the location width. Several variables
of the same type can be listed together, like:
double x, y, z;

identifying three storage locations for variables of type double.

To be useful, data needs to be stored in these locations. This is done with an assignment statement. For example:
width = 5;

A simple schematic diagram with a name for a location in memory (the box):

Although we are used to reading left to right, an assignment statement works right to left. The value on the right side
of the equal sign is calculated and then placed in the memory location associated with the variable on the left side of
the equal sign, either giving an initial value or overwriting any previous value stored there.

Variables can also be initialized as they are declared:

18 Chapter 2. C# Data and Operations

 Introduction to Computer Science in C#, Release 2.0

int width = 5;
double x = 12.5, y = 27, z = 0.5;

or initializations and plain declarations can be mixed:

int width = 5, height, area;
height = 7;

Stylistically the example above is inconsistent, but it illustrates what is possible. Technically an initialization is not an
assignment. We will see some syntax that is legal in initializers, but not declarations.
We could continue with a further assignment statement:
area = width * height;

Look at this in detail. The assignment statement starts by evaluating the expression on the right-hand side: width
* height. When variables are used in an expression, their current values are substituted, like in evaluating an
expression in math, so the value is the same as
5*7
which finally evaluates to 35. In the last step of the assignment statement, the value 35 is then assigned to the variable
on the left, area.

Warning: You want one spot in memory prepared for each variable. This happens with declaration, not assign-
ment: Assignment just changes the value at the current location. Do not declare the same variable more than once.
You will get an error. More on the fine points around that in Local Scope.

We continue introducing Csharp: Remember that in csharp you can just give an expression (and a semicolon), and
csharp responds with a value. That syntax and reaction is special to csharp. In csharp you can also test regular C#
statements, like declarations and assignments. As in a regular program, statements do not give an immediate visible
response in csharp. Still in csharp you still can display a variable value easily:
csharp> int width = 5, height, area;
csharp> height = 7;
csharp> area = width * height;
csharp> area;
35

In the last line, area is an expression, and csharp will give back its value, which is just the current value of the
variable.
At this point you should be able to make sense of some more features of csharp. You can start with the csharp special
help command:
csharp> help;
"Static methods:
Describe (object) - Describes the object’s type
LoadPackage (package); - Loads the given Package (like -pkg:FILE)
LoadAssembly (assembly) - Loads the given assembly (like -r:ASSEMBLY)
ShowVars (); - Shows defined local variables.
ShowUsing (); - Show active using declarations.
Prompt - The prompt used by the C# shell
ContinuationPrompt - The prompt for partial input
Time(() -> { }) - Times the specified code
print (obj) - Shorthand for Console.WriteLine
quit; - You’ll never believe it - this quits the repl!
help; - This help text
TabAtStartCompletes - Whether tab will complete even on emtpy lines

2.4. Variables and Assignment 19

Introduction to Computer Science in C#, Release 2.0

A lot of this is still beyond us but these parts are useful:

ShowVars (); - Shows defined local variables.
quit; - You’ll never believe it - this quits the repl!
help; - This help text

We can continue the csharp session above and illustrate ShowVars():

csharp> ShowVars();
int width = 5
int height = 7
int area = 35

displaying all the variables currently known to csharp, plus their current values.
We refer to “current values”. An important distinction between variables in math and variables in C# is that C# values
can change. Follow this csharp sequence:
csharp> int n = 3;
csharp> n;
3
csharp> n = 7;
csharp> n;
7

showing we can change the value of a variable. The most recent assignment is remembered (until the next assign-
ment....) We can imagine a schematic diagram:

We can carry this csharp session one step further, illustrating a difference between C# and math:
csharp> n = n + 1;
csharp> n;
8

Clearly n = n + 1 is not a true mathematical equation: It is a C# assignment, executing with a specific sequence of
steps.
1. First the right hand side expression is evaluated, n + 1.
2. This involves looking up the current alue of n, which we set to 7, so the expression is the same as 7 + 1 which
is 8.
3. After this evaluation, an assignment is made to the left hand variable, which happens to be n again.
4. Then the new value of n is 8, replacing the old 7.
There are many occasions in which such an operation will be useful.
Assignment syntax does have two strikes against it:
1. It appropriates math’s equal sign to mean something quite different.
2. The right to left operation is counter to the English reading direction.
Still this usage is common to many programming languages.

Warning: Remember in an assignent that the sides of the equal sign have totally different meanings. You assign
to a variable on the left side after evaluating the expression on the right.

We can illustrate a likely mistake in csharp:

20 Chapter 2. C# Data and Operations

 Introduction to Computer Science in C#, Release 2.0

csharp> 3 = n;
{interactive}(1,2): error CS0131: The left-hand side of an assignment
must be a variable, a property or an indexer

Students commonly try to assign left to right. At least in this case you get an error message so you see a mistake. If
you mean to assign the value of x to y, and write:
x = y;

you get the opposite effect, changing x rather than y, with no error statement. Be careful!
There is some weirdness in csharp because it adds special syntax for expressions that does not appear in regular
programs, but it also wants to allow syntax of regular programs. Some conflict can occur when trying to display
an expression, sometimes leading to csharp giving a strange error for apparently no reason. In that case, try putting
parentheses around the expression:
csharp> int width = 3;
csharp> int height = 5;
csharp> width * height;
{interactive}(1,2): error CS0246: The type or namespace name ’width’ could
not be found. Are you missing a using directive or an assembly reference?
csharp> (width * height);
15

2.4.1 Literals and Identifiers

Expressions like 27 or 32.5 or "hello" are called literals, coming from the fact that they literally mean exactly
what they say. They are distinguished from variables, who value the compiler cannot infer directly from the name
alone.
The sequence of characters used to form a variable name (and names for other C# entities later) is called an identifier.
It identifies a C# variable or other entity.
There are some restrictions on the character sequence that make up an identifier:
• The characters must all be letters, digits, or underscores _, and must start with a letter. In particular, punctuation
and blanks are not allowed.
• There are some words that are keywords for special use in C#. You may not use these words as your own
identifiers. They are easy to recognize in editors like in Xamarin Studio, that know about C# syntax: They are
colored differently.
We will only discuss a small fraction of the keywords in this course, but the curious may look at the full list.
C# is case sensitive: The identifiers last, LAST, and LaSt are all different. Be sure to be consistent. The compiler
can usually catch these errors, since it is the version used in the one declaration that matters.
What is legal is distinct from what is conventional or good practice or recommended. Meaningful names for variables
are important for the humans who are looking at programs, understanding them, and revising them. That sometimes
means you would like to use a name that is more than one word long, like price at opening, but blanks are
illegal! One poor option is just leaving out the blanks, like priceatopening. Then it may be hard to figure out
where words split. Two practical options are
• underscore separated: putting underscores (which are legal) in place of the blanks, like price_at_opening.
• using camel-case: omitting spaces and using all lowercase, except capitalizing all words after the first, like
priceAtOpening

2.4. Variables and Assignment 21

Introduction to Computer Science in C#, Release 2.0

Use the choice that fits your taste (or the taste or convention of the people you are working with). We will tend to use
camel-case for variable inside programs, while we use underscores in program file names (since different operating
systems deal with case differently).

Assignment Exercise

Think what the result would be in csharp:

int x = 1;
x = x + 1;
x = x * 3;
x = x * 5;
x;

Write your prediction. Then test. Can you explain it if you got it wrong?

2.5 Syntax Template Typography

When new C# syntax is introduced, the usual approach will be to give both specific examples and general templates.
In general templates for C# syntax the typeface indicates the the category of each part:
Typeface Meaning
Typewriter font Text to be written verbatim
Bold A place where you can use an arbitrary identifier.
Emphasized A place where you can use an arbitrary expression (which might be a single variable name).
Normal text A description of what goes in that position, without giving explicit syntax
An attempt is made with the parts that are not verbatim to be descriptive of the use expected.
As a start we can give some general syntax for declarations and assignment statements:

2.5.1 Declaration Syntax Options

type variableName ;
or with initialization:
type variableName = initialValue ;
or there can be a list of variables of the same type, for instance a list of three variables:
type variableName1 , variableName2 , variableName3 ;
Some or all of the variables in the list could also have initializers.
Space is allocated for each variable named, according to its type. Where there is an initializer, an initial value is set
for the variable.

2.5.2 Assignment Syntax

variableName = expression ;
The expression is evaluated before its value is assigned to variableName.

22 Chapter 2. C# Data and Operations

 Introduction to Computer Science in C#, Release 2.0

2.6 Strings, Part I

Enough with numbers for a while. Strings of characters are another important type in C#.
A string in C# is a sequence of characters. For C# to recognize a literal sequence of characters, like hello, as a
string, it must be enclosed in quotes " to delimit the string. Special cases are considered later in String Special Cases.

2.6.1 String Concatenation

We can operate on numbers with arithmetic operators, including +. Because everything in C# is typed, C# can give a
special meaning to operators depending on the types involved. This means + can have a special meaning with a string.
Look at the example in csharp:
csharp> "very " + "hot";
"very hot"

The plus operation with strings means concatenate the strings: join them together end to end.
C# is even a bit smarter. If you use a + with a string, presumable you are looking to produce a string, so even if the
second operand to the + is not a string, it is automatically converted to a string representation before concatenating:
csharp> int x = 42;
csharp> string result;
csharp> result = "We get " + x;
csharp> result;
"We get 42"

You can chain concatenations. We could make a full sentence adding a period:
csharp> "We get " + x + ".";
"We get 42."

Four Copies Exercise

In csharp declare and initialize a string variable. Write an expression that evaluates to four copies of the string, so it
work no matter with value you gave your string.

Thirty-two Copies Exercise

This is an extension of the previous exercise, except with 32 copies, but do not do it with one long expression. Include
some extra short assignment statements in the middle, to shorten the overall writing. Hint: 32 was chosen since you
can reach it by repeated doubling. To repeatedly double, you must save the result after each intermediate doubling.

Sum String Exercise

In csharp declare and initialize two int’s, x and y. Then enter an expression whose value is “x + y is 56”, except that 56
is replaced by the sum of x and y, and is not a literal, but calculated from the actual values of variables x and y (which
do not need to add up to 56 specifically).
This has a trick to it.

2.6. Strings, Part I 23

Introduction to Computer Science in C#, Release 2.0

Ints and Strings Added

In csharp enter
int x = 2;
int y = 3;

Think what the csharp response is to each of these, write one predicted response at a time, then then test it, and put the
right answer beside it if you were wrong:
x + "??" + y;
x + y + "??";
(x * y + "??");
"??" + x * y;
x + "??" * y;

Can you explain the ones you got wrong, after looking at the actual answer?

2.7 Writing to the Console

In interactive use of csharp, you can type an expression and immediately see the result of its evaluation. This is fine to
test out syntax and maybe do simple calculator calculations. In a regular C# program run from a file like in A Sample
C# Program, you must explicitly give instructions to print to a console or terminal window. This will be a window
like you see when running csharp.
This printing is accomplished through a function with a long name: Console.WriteLine. Like with math, you
can pass a function a value to work on, by placing it in parentheses after the name of the function. Unlike in high
school algebra classes, in C# we have many types of data to supply other than numbers. The simplest way to use the
Console.WriteLine function is to give it a string. We can demonstrate in csharp. The response is just the line
that would be printed in a regular program:
csharp> Console.WriteLine("Hello, world!");
Hello, world!

What is printed to the screen does not have the quotes we needed to define the literal string inside the program.
Console is a C# class maintained by the system, that interacts with the terminal or console window where text output
appears for the program. A function defined in that class is WriteLine. To refer to a function like WriteLine in a
different class, you must indicate the location of the function with the “dot” notation shown: class name, then ., then
the function. This gives the more elaborate name needed in the program.
The string that gets printed can be the result of evaluating an expression, for instance concatenating:
csharp> int total = 5;
csharp> Console.WriteLine("All together: " + total);
All together: 5

More elaborate use of WriteLine is discussed in String Format Operation.

The Console.WriteLine function automatically makes the printing position advance to the next line, as when
you press the Enter or Return key. A variant, Console.Write, prints the parameter exactly, and nothing else. The
statement-at-a-time approach in csharp is not good for illustrating the differences.
Printing is better shown off in a real program....

24 Chapter 2. C# Data and Operations

 Introduction to Computer Science in C#, Release 2.0

2.8 C# Program Structure

We discuss the most basic syntax satisfied by all C# programs, which are plain text files, with names ending in .cs.
There will be additions later, but any program you can run will include:
using System;

class ClassName
{
static void Main()
{
program statements go here....

}
}
By convention class names are capitalized.
You can see that both the example program painting/painting.cs and the lab program hello/hello.cs follow this pattern.
The discussion of these parts through line 6 in A Sample C# Program are about all we have to say at this point. For
now this is the boilerplate code. We will make additions as necessary. We choose not to clutter up the basic setup with
features that we are not about to use and discuss.
Here is a silly little test illustrating the difference between Console.WriteLine and Console.Write, in ex-
ample write_test/write_test.cs:
When run, the program prints:
hellotherehello
Another line
Starting yet another line

Do you see how the output shows the differences between WriteLine and Write? If we added another printing
statement, where would the beginning of the output appear: after the final e or under the S of Starting?

2.9 Combining Input and Output

2.9.1 Reading from the Keyboard

If you want users to type something at the keyboard, you should let them know first! The jargon for this is to give
them a prompt: Instructions written to the screen, something like
Console.Write("Enter your name: ");

Then the user should respond. What the user types is automatically shown (echoed) in the terminal or console window.
For a program to read what is typed, another function in the Console class is used: Console.ReadLine.
Here the data for the function comes from a line typed at the keyboard by the user, so there is no need for a parameter
between the parentheses: Console.ReadLine(). The resulting sequence of characters, typed before the user
presses the Enter (Return) key, form the string value of the function. Syntactically in C#, when a function with a value
is used, it is an expression in the program, and the expression evaluation is the value produced by the function. This is
the same as in normal use of functions in math.
With any function producing a value, the value is lost unless it is immediately used. Often this is done by immediately
assigning the value to a variable like in

2.8. C# Program Structure 25

Introduction to Computer Science in C#, Release 2.0

string name;
name = Console.ReadLine();

or in the shorter
string name = Console.ReadLine();

Fine point: Notice that in most operating systems you can edit and correct your line before pressing the Return key.
This is handy, but it means that the Return key must always be pressed to signal the end of the response. Hence
a whole line must be read, and there is no function Console.Read(). Just for completeness we mention that
you can read a raw single keystroke immediately (no editing beforehand). If you want to explore that later, see
test_readkey/test_readkey.cs.

2.9.2 Numbers and Strings of Digits

You may well want to have the user supply you with numbers. There is a complication. Suppose you want to get
numbers and add them. What happens with this code, in bad_sum/bad_sum.cs?
Here is a sample run:
Enter an integer: 23
Enter another integer: 9
They add up to 239

C# has a type for everything and Console.ReadLine() gives you a string. Adding strings with + is not the same
as adding numbers!
We must explicitly convert the strings to the proper kind of number. There are functions to do that: int.Parse takes
a string parameter that should be the characters in an int, like “123” or “-25”, and produces the corresponding int
value, like 123 or -25. In good_sum/good_sum.cs, we changed the names to emphasize the type conversions:
Notice that the values calculated by int.Parse for the strings xString and yString are immediately remem-
bered in assignment statements. Be careful of the distinction here. The int.Parse function does not magically
change the variable xString into an int: the string xString is unchanged, but the corresponding int value is
calculated, and gets assigned to an int variable, x.
Note that this would not work if the string represents the wrong kind of number, but there is an alternative:
csharp> string s = "34.5";
csharp> int.Parse(s);
System.FormatException: Input string was not in the correct format ....
csharp> double.Parse(s);
34.5

We omitted the long tail of the error message. There is no decimal point in an int. You see the fix with the corre-
sponding function that returns a double.

2.9.3 Example Projects and the Source Repository

We have started to refer to whole programs that we have written. You will want to have your own copies to test and
modify for related work.
All of our examples are set up in a Xamarin Studio solution in our . zip file that you can download.
Unzip source.zip in a place you control. The examples subdirectory is a Xamarin Studio solution.
There are various way to access our files.
1. One way is to look at individual files from your download under our examples directory.

26 Chapter 2. C# Data and Operations

 Introduction to Computer Science in C#, Release 2.0

2. If you open the examples solution in Xamarin Studio, you can select files from the Solutions pad.
3. In the notes we refer to individual code file names that are hyperlinked. They link to the latest version in our
online source repository. You get a display of color-coded web page with numbered lines. If you want to adapt
a chunk, you can select it, and copy. If you want to copy all of a large file, your editing shortcuts for Select All
do not work: You get a bunch of extra html. An alternative is to click the Raw button in the web page to the left
above the source code. That brings up a plain text page with just the code. You can either download it or select
all of it, and you only get the code.
We have one main convention in naming our projects: Most projects are examples of full, functional programs to run.
Others are intended to be copied by you as stubs for your solutions, for you to elaborate. These project folders all
end with “_stub”, like string_manip_stub. Even the stubs can be compiled immediately, though they may not
accomplish anything.

2.9.4 Running our Xamarin Studio Examples Solution

If you are just starting Xamarin Studio, and you have not run our solution before:
1. On the Welcome screen select the button Open Solution or File.
2. You get anopen-file dialog. Navigate to our example solution. (It must be unzipped already!)
3. Select examples/examples.sln.
The next time you come to the Welcome screen, our examples should be listed in the Recent Projects, and you can
click to open it directly.

2.9.5 Copying and Modifying Our Example Xamarin Studio Projects

We strongly encourage you not to modify our examples in place, if you want to keep the changes, because we will
make additions and modifications to source.zip, and the easiest thing to do is to unzip the new version on top of the
old version, clobbering any changes that you made to files.
If you do want to alter our code, we suggest you copy it to a project in your solution (“work”, discussed before).
1. Open your solution.
2. Create a new project, maybe with the same name as the one we had. If it was a “_stub” project, remove the
“_stub” from your project’s name.
3. In the Solution Pad open the menu on the new project, select, Add, and then in the further submenu, select Add
Files....
4. This brings up an operating system open-file dialog. Switch folders into our example projects. Select the files
you want to copy.
5. A further dialog window pops up, with the choice Copy selected. Click to approve copy (as opposed to move
or link).
6. Now the desired files should appear in your project. If you intended to copy everything for a project, test by
running the project. Even our stub projects should compile, though a stub project may not do anything when
you run it until you add your own code to it. To make successful incremental additions, it is always good to start
from something the compiles!

Exercise for Addition

Write a version, add3.cs, that prompts the user for three numbers, not necessarily integers, and lists all three, and
their sum, in similar format to good_sum/good_sum.cs.

2.9. Combining Input and Output 27

Introduction to Computer Science in C#, Release 2.0

Exercise for Quotients

Write a program, quotient.cs, that prompts the user for two integers, and then prints them out in a sentence with
an integer division problem like
The quotient of 14 and 3 is 4 with a remainder of 2

Review Division and Remainders if you forget the integer division or remainder operator.

2.10 String Special Cases

There are some special cases for creating literal strings. For instance you might want quotes as characters inside
your string. In this case you need special symbolism using a string escape code, starting with \ backslash. Then the
character after the backslash has a special meaning.
For instance a quote character after a backslash, \", does not mean the end of the string. It means a quote character is
literally used in the string: "He said, \"Hello!\", over and over."
We can illustrate with csharp, first with a simple string:
csharp> Console.WriteLine("Hello world!");
Hello world!
csharp> Console.WriteLine("He said, \"Hello!\", over and over.");
He said, "Hello!", over and over.

There are many other special cases of escape code. The main ones you are likely to use are:
Escape code Meaning
\" " (quote)
\\ \ (backslash)
\n newline
Hence if you really want a backslash character in a literal string, you need to write two of them.
The newline character indicates further text will appear on the next line down when printed with the
Console.WriteLine function.
Example:
csharp> Console.WriteLine("Windows path: c:\\Users\\aharrin");
Windows path: c:\Users\aharrin
csharp> Console.WriteLine("a\nbc\n\ndef")
a
bc

def

Literal strings that are simply delimited by quotes " must start and end on the same line. There is also a notation
for @-quoting, with an at-sign @ before the first quote. In an @-quoted string, all characters are treated literally,
including all backslashes. Also the string may go on for several lines, and all newlines are included literally. (The
csharp program does not recognize multi-line @-quoted strings.) This fragment in a program would produce the same
output as the statements in the csharp example above:
Console.WriteLine(@"Windows path: c:\Users\aharrin");
Console.WriteLine(@"a
bc

def");

28 Chapter 2. C# Data and Operations

 Introduction to Computer Science in C#, Release 2.0

The only thing this example does not show off well is the amount of left margin indentation. That is significant in a
multiline @-quoted string. A whole simple program with this code is in example at_sign_strings/at_sign_strings.cs.

2.11 Substitutions in Console.WriteLine

2.11.1 Output With +

An elaboration of a “Hello, World” program, could greet the user, after obtaining the user’s name. If the user enters
the name Kim, the program could print
Hello, Kim!
This is a very simple input-process-output program (in fact with almost no “process”). Think how would you code it.
You need to obtain a name, remember it and use it in your output. A solution is in the next section.

2.11.2 String Format Operation

A common convention is fill-in-the blanks. For instance,

Hello, _____!
and you can fill in the name of the person greeted, and combine given text with a chosen insertion. C# has a similar
construction, better called fill-in-the-braces, that can be used with Console.WriteLine.
Instead of inserting user input with the + operation as in hello_you1/hello_you1.cs:
look at a variation, hello_you2/hello_you2.cs, shown below. Both programs look exactly the same to the user:
All the new syntax is in the line:
Console.WriteLine ("Hello, {0}!", name);

Console.WriteLine actually can take parameters after an initial string, but only when the string is in the form of
a format string, with expression(s) in braces where substitutions are to be made, (like in fill-in-the-blanks). Here the
format string is "Hello, {0}!".
The remaining parameters, after the initial string, give the values to be substituted. To know which further parameter to
substitute, the parameters after the initial string are implicitly numbered, starting from 0. Starting with 0 is consistent
with other numbering sequences in C#. So here, where there is just one value to substitute (name), it gets the index 0,
and where it is substituted, the braces get 0 inside, to indicate that parameter with index 0 is to be substituted.
Everything in the initial string that is outside the braces is just repeated verbatim. In particular, if the only param-
eter is a string with no braces, it is printed completely verbatim (reducing to the situations where we have used
Console.WriteLine before).
A more elaborate silly examples that you could test in csharp would be:
string first = "Peter";
string last = "Piper";
string what = "pick";
Console.WriteLine("{0} {1}, {0} {1}, {2}.", first, last, what);

It would print:
Peter Piper, Peter Piper, pick.

2.11. Substitutions in Console.WriteLine 29

Introduction to Computer Science in C#, Release 2.0

where parameter 0 is first (value "Peter"), parameter 1 is last ( value "Piper"), and parameter 2 is what
(value "pick").
Make sure you see why the given output is exactly what is printed.
Or try in csharp:
int x = 7;
int y = 5;
Console.WriteLine("{0} plus {1} is {2}; {0} times {1} is {3}.", x, y, x+y, x*y);

and see it print:

7 plus 5 is 12; 7 times 5 is 35.

Note the following features:

• Parameters can be any expression, and the expressions get evaluated before printing.
• Parameters to be substituted can be of any type.
• The parameters are automatically converted to a string form, just as in the use of the string + operation.
In fact the simple use of format strings shown so far can be completed replaced by long expressions with +, if that is
your taste. We later discusses fancier formatting in Tables, that cannot be duplicated with a simple string + operation.
We will just use the simple numbered substitutions for now, to get used to the idea of substitution.
A technical point: Since braces have special meaning in a format string, there must be a special rule if you want braces
to actually be included in the final formatted string. The rule is to double the braces: "{{" and "}}". The fragment
int a = 2, b = 3;
Console.WriteLine("The set is {{{0}, {1}}}.", a, b);

produces
The set is {2, 3}.

2.11.3 Overloading

The WriteLine function can take parameters in different ways:

• It can take a single parameter of an type (and print its string representation).
• It can take a string parameter followed by any number of parameters used to substitute into the initial format
string.
• It can take no parameters, and just advance to the next line (not used yet in this book).
Though each of these uses has the same name, Console.WriteLine, they are technically all different functions:
A function is not just recognized by its name, but by its signature, which includes the name and the number and
types of parameters. The technical term for using the same name with different signatures for different functions is
overloading the function (or method).
This only makes practical sense for a group of closely related functions, where the use of the same name is more
helpful than confusing.

30 Chapter 2. C# Data and Operations

 Introduction to Computer Science in C#, Release 2.0

2.12 Value Types and Conversions

2.12.1 Type int

A variable is associated with a space in memory. This space has a fixed size associated with the type of data. The
int and double types are examples of value types, which means that this memory space holds an encoding of the
complete data for the value of the variable. The fixed space means that an int cannot be a totally arbitrary integer of
an enormous size. In fact an int variable can only hold an integer in a specific range.
The smallest memory unit in a modern computer is a bit with two states, that can be thought of as 0 or 1. An
int is held in a memory space of 32 bits, so it can have at most 232 values, and the encoding is chosen so about
half are positive and half are negative: An int has maximum value 231 − 1 = 2147483647 and a minimum value of
−231 = −2147483648. The extreme values are also named constants in C#, int.MaxValue and int.MinValue.
In particular this means int arithmetic does not always work. What is worse, it fails silently:
csharp> int i = int.MaxValue;
csharp> i;
2147483647
csharp> i + 5;
-2147483644

Add two positive numbers and get a negative number! This is called overflow. Be very careful if you are going to be
using big numbers!

2.12.2 Type long

Most everyday uses of integers fit in this range of an int, and modern computers are designed to operate on an int
very efficiently, but sometimes you need a larger range. Type long uses twice as much space.
The same kind of silent overflow errors happen with long arithmetic, but only with much larger numbers.
When we get to Arrays, you will see that a program may store an enormous number of integers, and then the total
space may be an issue. If each of these numbers fit in a more modest range of values, they can be stored in the smaller
space of a short. There are other smaller types, too. We will not have need for integral types other than int and
long in this book.

2.12.3 Type double

A double is also a value type, stored in a fixed sized space. There are even more issues with double storage than
with an int: Not only do you need to worry about the total magnitude of the number, you also need to choose a
precision: There are an infinite number of real values, just between 0 and 1. Clearly we cannot encode for all of them!
As a result a double has a limited number of digits of accuracy. There is also an older type float that takes up
half of the space of a double, and has a smaller range and less accuracy. This at least gives a reason for the name
double: double the storage space of a float.
To avoid a ridiculously large number of trailing 0’s, a big double is expressed using a variant of scientific notation:
1.79769313486232E+308 means 1.7976931348623157(10308 )
C# does not have the typography for raised exponents. Instead literal values can use the E to mean “times 10 to the
power”, and the E is followed by and exponent integer that can be positive or negative. The whole double literal may
not contain any embedded blanks.

2.12. Value Types and Conversions 31

Introduction to Computer Science in C#, Release 2.0

The lack of error handling with type int is not repeated with doubles. We show behavior that could be important if
you do scientific computing with enormous numbers: Arithmetic with the double type does not overflow silently as
with the integral types. There are values for infinity and minus infinity and Not a Number (NaN):
csharp> double x = double.MaxValue;
csharp> x;
1.79769313486232E+308
csharp> double y = 10 * x;
csharp> y;
Infinity
csharp> y + 1000;
Infinity
csharp> y - 1000;
Infinity
csharp> 1000/y;
0
csharp> double z = 10 - y;
csharp> z;
-Infinity
csharp> double sum = y + z;
csharp> sum;
NaN
csharp> sum/1000;
NaN

Once a result gets too big, it gets listed as infinity. As you can see, there is some arithmetic allowed with a finite
number and infinity! Still some operations are not legal. Once a result turns into NaN, no arithmetic operations change
further results away from NaN, so there is a lasting record of a big error!
There is no such neat system for showing off small inaccuracies in double arithmetic accumulating due to limited
precision. These inaccuracies still happen silently.

2.12.4 Numeric Type Limits

The listing below shows how the storage size in bits translates into the limits for various numerical types. We will not
discuss or use short or float further.
long 64 bits; range -9,223,372,036,854,775,808 to 9,223,372,036,854,775,807
int 32 bits; range -2,147,483,648 to 2,147,483,647
short 16 bits; range -32,768 to 32,767
double 64 bits; maximum magnitude: 1.7976931348623157(10308 ); about 15 digits of accuracy
float 32 bits; maximum magnitude: 3.402823(1038 ); about 7 digits of accuracy

2.12.5 Type char

The type for an individual character is char. A char literal value is a single character enclosed in single quotes, like
’a’ or ’$’. The literal for a single quote character itself and the literal for a newline use escape codes, like in String
Special Cases: The literals are ’\’’ and ’\n’ respectively.
Be careful to distinguish a char like ’A’ from a string "A".
An individual character is also technically a number, with the correspondence between numeric codes and characters
given by the Unicode standard. Unicode allows special symbol characters and alphabets of many languages. We will
stick to the standard American keyboard for these characters.

32 Chapter 2. C# Data and Operations

 Introduction to Computer Science in C#, Release 2.0

2.12.6 Casting

While the mathematical ideas of 42 and 42.0 are the same, C# has specific types. There are various places where
numerical types get converted automatically by C# or explicitly by the programmer. A major issue is whether the new
type can accurately represent the original value.
Going from int to double has no issue: Any int can be exactly represented as a double. Code like the following
is fine:
csharp> int i = 33;
csharp> double d = i;
csharp> double x;
csharp> x = 11;
csharp> double z = i + 2.5;
csharp> ShowVars();
int i = 33
double d = 33
double x = 11
double z = 35.5

The double variable d is initialized with the value of an int variable. The double variable x is assigned a value
using an int literal. The double variable z is initialized with the value of a sum involving both an int variable and
a double literal. As we have discussed before in Arithmetic, the int is converted to a double before the addition
operation is done.
The other direction for conversion is more problematic:
csharp> double d= 2.7;
csharp> int i;
csharp> i = d;
{interactive}(1,4): error CS0266: Cannot implicitly convert type ’double’ to ’int’.
An explicit conversion exists (are you missing a cast?)

The int i cannot accurately hold the value 2.7. Since the compiler does this checking, looking only at types, not
values, this even fails if the the double happens to have an integer value:
csharp> double d = 2.0;
csharp> int i = d;
{interactive}(1,4): error CS0266: Cannot implicitly convert type ’double’ to ’int’.
An explicit conversion exists (are you missing a cast?)

If you really want to possibly lose precision and use a double to produce an int result, you can do it, but you must
be explicit, using a cast as the csharp error messages suggest.
csharp> double d= 2.7;
csharp> int i;
csharp> i = (int)d;
csharp> i;
2

The desired result type name in parentheses (int) is a cast, telling the compiler you really intend the conversion.
Look what is lost! The cast does not round to the nearest integer, it truncates toward 0, dropping the fractional part, .7
here.
Rounding is possible, but if you really want the int type, it takes two steps, because the function Math.Round does
round to a mathematical integer, but leaves the type as double! To round d to an int result we could use:
csharp> i = (int)Math.Round(d);
csharp> i;
3

2.12. Value Types and Conversions 33

Introduction to Computer Science in C#, Release 2.0

You can also use an explicit cast from int to double. This is generally not needed, because of the automatic conversions,
but there is one place where it is important: If you want double division but have int parts. Here is a quick artificial
test:
csharp> int sum = 14;
csharp> int n = 4;
csharp> double avg = sum/n;
csharp> avg;
3

Oops, integer division. Instead, continue with:

csharp> avg = (double)sum/n;
csharp> avg;
3.5

We get the right decimal answer.

This is a bit more subtle than it may appear: The cast to double, (double) is an operation in C# and so it has
a precedence like all operations. Casting happens to have precedence higher than any arithmetic operation, so the
expression is equivalent to:
avg = ((double)sum)/n;

On the other hand, if we switch the order the other way with parentheses around the division:
csharp> avg = (double)(sum/n);
csharp> avg;
3

then working one step at a time, (sum/n) is integer division, with result 3. It is the 3 that is then cast to a double (too
late)!
See the appendix Precedence of Operators, listing all C# operations discussed in this book.

2.12.7 Integral Type char

Though the char type has character literals and prints as a character, internally a char is a type of integer, stored in
16 bits. We mention the char type being numeric mostly because of errors that you can make that would otherwise
be hard to figure out. This code does not concatenate:
csharp> Console.WriteLine(’A’ + ’-’);
110

What? We mentioned that modern computers are set up to easily work with the int type. In arithmetic with smaller
integral types the operands are first automatically converted to type int. An int sum is an int, and that is what is
printed.
You can look at the numeric values inside a char with a cast!
csharp> int n = (int)’A’;
csharp> n;
65
csharp> int m = (int)(’-’);
csharp> m;
45

So the earlier 110 is correct: 65 + 45 = 110.

For completeness: It is also possible to cast back to char. This may be useful for dealing with the alphabet in sequence
(or simple classical cryptographic codes):

34 Chapter 2. C# Data and Operations

 Introduction to Computer Science in C#, Release 2.0

csharp> char ch = (char)(’A’ + 1);

csharp> ch;
’B’

The capital letter one place after A is B.

2.13 Learning to Solve Problems

This section might have been placed earlier, but by placing it here, you should realize that you will have a lot of data
and concepts to deal with.
The manner in which you deal with all the data and ideas is very important for effective learning. It might be rather
different than what you needed if you were in a situation where rote recall is the main important thing.
Different learning styles mean different things are useful to different people. Consider what is mentioned here and try
out some approaches.
The idea of this course is not to regurgitate the book, but to learn to solve problems (generally involving producing a
computer program). In this highly connected and wired world you have access to all sorts of data. The data is not an
end in itself, the question is doing the right things with the tools out there to solve a new creative problem.
In this course there is a lot of data tied into syntax and library function names and .... It can seem overwhelming. It
need not be. Take a breath.
First basic language syntax: When learning any new language, there is a lot to take in. We introduce C# in chunks. For
a while there will always be the new current topic coming. You do NOT need to memorize everything immediately!
• Some things that you use rarely, you may never memorize, like, “What is the exact maximum magnitude of a
double?” At some point that might be useful. Can you find it? It happens to be in Numeric Type Limits. It is
also in online .Net documentation that you can Google or bookmark.
• Some things you will use all the time, but of course they start off as new and maybe strange. Knowing where
to go to check is still useful but not sufficient. For much-used material that you do not find yourself absorbing
immediately, consider writing down a summary of the current topic. Both thinking of a summary and writing
help reinforce things and get you to remember faster. Also, if you have the current things of interest summarized
in one place, they are easy to look up!
• If you need some syntax to solve a simple early problem, first try to remember the syntax, then check. With
frequently used material and with this sort of repetition, most everyone will remember most everything shortly.
If there are a few things that just do not stick, keep them in your list. Then go on to new material. The list of
what you need to check on will keep changing as you get more experience and get to more topics. If you keep
some of the old lists, you will be amazed how much stuff that you sweated over, is later ho-hum or automatic.
• In the earliest exercises the general steps that you need should be pretty apparent, and you can just concentrate
on translating simple ideas into C# syntax (mostly from the material most recently introduced). In this case the
focus is mostly on syntax.
Memorizing syntax is not going to directly get you to solve real problems. In any domain: programming, construction,
organizing political action, ..., you need to analyze the problem and figure out a sequence of steps, knowing what
powers and resources you have.
For example with political action: if you know demonstrations are possible in front of City Hall, you can make a
high-level plan to have one, but then you have to attend to details: Do you need city permission? Who do you call? ...
You do not have to have all that in your head when coming up with the idea of the demonstration, but you better know
how to find the information allowing you to follow through to make it happen.
With programming, syntax details are like the details above: not the first thing to think of, and maybe not things that
you have memorized. What is important to break down a problem and plan a solution, is to know the basic capacities

2.13. Learning to Solve Problems 35

Introduction to Computer Science in C#, Release 2.0

you have in programming. As you get into larger projects and have more experience, “basic capacities” will be bigger
and bigger ideas. For now, as beginners, it is important to know:
• You can get information from a user and return information via keyboard and screen.
• You can remember and recall and use information using variables.
• You can deal directly with various kinds of data: numbers and strings at this point.
• There are basic operations you can do with the data (arithmetic, concatenating string, converting between data
types).
• At a slightly higher level, you might already have the idea of basic recurring patterns, like solving a straightfor-
ward problem with input-processing-output.
• You will shortly see that you have more tools: decision, repetition, more built-in ways to deal with data (like
more string operations shortly), creating your own data types....
At slightly more detailed level, after thinking of overall plans:
• There are multiple kinds of number types. What is appropriate for your use?
• There are various ways of formatting and presenting data to output. What shall you use?
Finally, you actually need to translate specific instructions into C# (or whatever language). Of course if you remember
the syntax, then this level of step is pretty automatic. Even if you do not remember, you have something very specific
to look up! If you are keeping track of your sources of detailed information, this is hopefully only one further step.
Contrast this last-step translation with the earlier creative organizational process: If you do not have in your head an
idea of the basic tools available, how are you going to plan? How are you going to even know how to start looking
something up?
So far basic ideas for planning a solution has been discussed, and you can see that you do not need to think of
everything at once or have everything equally prominent in your brain.
Also, when you are coding, you do not need to to have all the details of syntax in your head, even for the one instruction
that you are dealing with at the moment. You want to have the main idea, and you want to get it written down, but
once it is written down, you can make multiple passes, examining and modifying what you have. For example, Dr.
Harrington does a lot of Python programming, where semicolons are not needed. He can get the main ideas down in
C# without the required semicolons. He could wait for the compiler to stop him on every one that is missed, and maybe
have the compiler misinterpret further parts, and give bogus error messages. More effective is having a list of things to
concentrate on in later rounds of manual checking. For example, checking for semicolons: Scan the statements; look
at the ends; add semicolons where missing. You can go through a large program very quickly and efficiently doing
this and have one less thing to obsess about when first writing.
This list of things-to-check-separately should come from experience. Keep track of the errors you make. Some people
even keep an error log. What errors keep occurring? Make entries in things-to-check-separately, so you will make
scans checking for the specific things that you frequently slip up on.
This things-to-check-separately list, too, will evolve. Revise it occasionally. If Dr. Harrington does enough con-
centrated C#, maybe he will find that entering semicolons becomes automatic, and he can take the separate round of
semicolon checking off his list....
What to do after you finish an exercise is important, too. The natural thing psychologically, particularly if you had a
struggle, is to think, “Whew, outta here!!!!”
On something that came automatically or flowed smoothly, that is not a big deal - you will probably get it just as fast
the next time. If you had a hard time and only eventually got to success, you may be doing yourself a disservice with
“Whew, outta here!!!”
We have already mentioned how not everything is equally important, and some things are more important to keep in
your head than others. The same idea applies to all the steps in solving a possibly long problem. Some parts were
easy; some were hard; there may have been many steps. If all of that goes into your brain in one continuous stream of

36 Chapter 2. C# Data and Operations

 Introduction to Computer Science in C#, Release 2.0

stuff that you remember at the same level, then you are going to leave important nuggets mixed in with an awful lot
of unimportant and basically useless information, and have it all fade into oblivion, or be next to impossible to cycle
through looking for the nuggets. Why do the problem anyway if you are just going to bury important information
further down in your brain?
What is important? The most obvious thing you will need at a higher level of recall is what just messed you up, what
you missed until doing this problem: After finishing the actual problem, actively follow up and ask yourself:
• What did I get in the end that I was missing initially? What was the connection I made?
• Does this example fit in to some larger idea/abstraction/generalization in a way that I did not see before?
• How am I going to look at this so I can make a similar connection in a similar (or maybe only partly similar)
problem?
• Is there a kernel here that I can think of as a new tool in my bag of tricks?
Your answers to these questions are the most important things to take away from your recent hard work. The extra
consideration puts them more in the “priority” part of your brain, so you can really learn from your effort. When you
need the important ideas next, you do not need to play through all the details of the stuff you did to solve the exact
earlier problem.
Keep coming back to this section and check up on your process: It is really important.

2.14 Lab: Division Sentences

2.14.1 Overview

In this lab, we’re going to begin to look at what makes computers do their thing so to speak.
It is rather insightful to look at how Wikipedia summarizes the computer:
A computer is a programmable machine designed to sequentially and automatically carry out a sequence
of arithmetic or logical operations. The particular sequence of operations can be changed readily, allowing
the computer to solve more than one kind of problem.
In other words, a computer is a calculator–and much more. Furthermore, the definition of a computer goes on to
include access to storage and peripherals, such as consoles (graphical displays), printers, and the network. We already
got a glimpse of this access when we explored Console.WriteLine in the first lab exercise.
So in this lab, we’re going to explore the use of C# as a calculator. We’re going to begin by looking at the csharp
command. Then we will take what we’ve learned in this session and use it to write a full program.

2.14.2 Requirements

We want to develop a program that can do the following:

• Prompt the user for input of two integers, which we will call numerator and denominator. For clarity, we are
only looking at integers, because this assignment is about rational numbers. A rational number can always be
expressed as a quotient of two integers.
• Calculate the floating point division result (e.g. 10/4 = 2.5).
• Calculate the quotient and the remainder (e.g. 10/4 = 2 with a remainder of 2 = 2 2/4).
Your final program should work as in this sample run, and use the same format:

2.14. Lab: Division Sentences 37

Introduction to Computer Science in C#, Release 2.0

Please enter the numerator? 14

Please enter the denominator? 4
Integer division result = 3 with a remainder 2
Floating point division result = 3.5
Result as a mixed fraction = 3 2/4

For this lab the example format 3 2/4 is sufficient. It would look better as 3 1/2, but a general efficient way to
reduce fractions to lowest terms is not covered until the section on the algorithm Greatest Common Divisor.
So let’s get this party started by firing up the csharp command, as introduced in Csharp and continued later with more
csharp. We can illustrate some of the ideas for this lab:
csharp> Console.WriteLine("Please enter the numerator?");
Please enter the numerator?
csharp> string input = Console.ReadLine();
25
csharp> int numerator = int.Parse(input);
csharp> Console.WriteLine(numerator);
25

Now let’s practice using the C# operators of Division and Remainders:

csharp> int quotient = numerator / denominator;
csharp> Console.WriteLine(quotient);
3
csharp> int remainder = numerator % denominator;
csharp> Console.WriteLine(remainder);
2
csharp> Console.WriteLine("{0} / {1} = {2} with a remainder {3}",
> numerator, denominator, quotient, remainder);
14 / 4 = 3 with a remainder 2

In the above, we are also practicing using a String Format Operation with Console.WriteLine.
You may find this example to be helpful to print the output according to the final lab requirements:
14 / 4 = 3 2/4

Now let’s take a look at how we can get the results as a real number, not necessarily an integer. Here is a variation
on the approach in Casting. We do this by declaring a couple of double variables to hold each of the numerator and
denominator integers. Then we will declare a variable to capture the result of the floating point division operation.
We named each of the floating-point variables with the number 2 in the name as C# permits variable names that have
numbers and underscores after the first character (which must be a letter or an underscore):
csharp> double numerator2 = numerator;
csharp> double denominator2 = denominator;
csharp> double quotient2 = numerator2/denominator2;
csharp> Console.WriteLine(quotient2);
3.5
csharp> Console.WriteLine("{0} / {1} = {2} remainder {3}",
> numerator, denominator, quotient, remainder);
14 / 4 = 3 remainder 2
csharp> Console.WriteLine("{0} / {1} = {2} approximately",
> numerator2, denominator2, quotient2);
14 / 4 = 3.5 approximately

We have shown everything you need to understand to complete this lab. To help you get started, we provided this
simple stub in the example file do_the_math_stub/do_the_math.cs.

38 Chapter 2. C# Data and Operations

 Introduction to Computer Science in C#, Release 2.0

The body of Main presently contains only comments, skipped by the compiler. We illustrate two forms (being incon-
sistent for your information only):
• // to the end of the same line
• /* to */ through any number of lines.
Save the stub in a project of your own and replace the comments with your code to complete it:
Be sure to run it and test it thoroughly. Show your output to a TA.

2.14.3 Xamarin Studio Reminders and Fixes

Be careful to open your Xamarin Studio solution and add a new C# Console project to it, and add your new file
directly into the project (through the Solution pad). There are two main places to mess up here. We emphasize them
and mention fixes if you make the easy mistakes:
1. It is easy to select Empty Project instead of C# and Console Project. If you do that, a correct program will
compile successfully, but it will run in limbo, with no console attached to it, and all Console.ReadLine()
calls return null, which is likely to make the program have a run-time error. One way to fix it:
• If you discovered this while running your program, there is no good access to the running process. (You
lack a console!) In this case you need to close your solution, ending the running process, and open the
solution again.
• Double click on the project in the Solution Pad (if that does anything, or right-click it and select Options).
An elaborate Project Options dialog window appears.
• In the left pane under Run, select General. In the right pane, two check boxes should appear. Make sure
you have the first checked: Run on external console. That should check the second one automatically.
Close the window and you should be set.
2. Another common error is to proceed like with most text processors, and open the top File menu, and choose
to open and edit a new file for your program. You cannot run this program from Xamarin Studio. If you have
a separate project set up, but without this file or any other showing in the Solutions pad, an attempt to run the
project with say no Main method (in fact no program at all). The fix:
• You will shortly need to navigate in an operating system open file dialog to where you put the file created
from the File menu. If you do not remember where that was, a good trick is to click in the edit window of
the file and then go to the File menu and select Save As. The dialog should show where the file currently
is. Cancel the dialog.
• Right click on the project in the Solution pad, and choose Add and then Add Files.... Browse to where the
file is and select it; click Open. Unless you have some reason to keep a copy in the original place, select
Move, and Ok. Now the orphaned file is moved into your project. You should see it list under the project
in the Solution pad. You can proceed to edit and run it.
3. If you lose the display of the Solution pad somehow, you can go to the View menu, select Pads, and then select
Solution.

2.14. Lab: Division Sentences 39

Introduction to Computer Science in C#, Release 2.0

40 Chapter 2. C# Data and Operations

 CHAPTER

THREE

DEFINING FUNCTIONS OF YOUR OWN

3.1 A First Function Definition

If you know it is the birthday of a friend, Emily, you might tell those gathered with you to sing “Happy Birthday to
Emily”.
We can make C# display the song. Read, and run if you like, the example program birthday1/birthday1.cs:
Here the song is just a part of the Main method that is in every program.
Note that we are using a function already provided to us, Console.WriteLine. We can use it over and over,
wherever we like. We can alter its behavior by including a different parameter. Now we look further at writing and
using your own functions.
If we want this song to be just part of a larger program, and be able to refer to it repeatedly and easily, we might like
to package it separately. You would probably not repeat the whole song to let others know what to sing. You would
give a request to sing via a descriptive name like “Happy Birthday to Emily”.
In C# we can also give a name like HappyBirthdayEmily, and associate the name with whole song by using a
new function definition, also called a method. We will see many variations on method definitions. Later we will see
definitions that are attached to a particular object. For now the simpler cases do not involve creating a type of object,
but there is an extra word needed to distinguish a function definition not attached to on object, static. We will also
shortly look at functions more like the functions from math class, that produce or return a value. In this simple case
we will not deal with returning a value. This also requires a special word in the heading: void. A void function will
just be a shorthand name for something to do, a procedure to follow, in this case printing out the Happy Birthday song
for Emily. (Note that the Main method for a program is also static void. This does your whole program and is
not attached to an object.)
Read for now:
There are several parts of the syntax for a function definition to notice:
Line 5: The heading starts with static void, the name of the function, and then parentheses.
A more general syntax for functions that just do something is
static void FunctionName()
{
statements in the function body...
}
Recall the conventions in Syntax Template Typography.
Lines 6-11: The remaining lines form the function body. They are enclosed in braces. By convention the lines inside
the braces are indented by a consistent amount. Three spaces is a common indentation.

41
Introduction to Computer Science in C#, Release 2.0

The whole definition does just that: defines the meaning of the name HappyBirthdayEmily, but it does not do
anything else yet - for example, the definition itself does not make anything be printed yet. This is our first example of
altering the order of execution of statements from the normal sequential order. This is important: the statements in the
function definition are not executed as C# first passes over the lines. The only part of a program that is automatically
executed is Main. Hence Main better refer to the newly defined function....
Look at the first statement inside Main, line 15:
HappyBirthdayEmily();

Note that the static void of the function definition is missing, but we still have the function name and parentheses.
When Main is running, C# goes back and looks up the definition, and only then, executes the code inside the function
definition. The term for this action is a function call or function invocation. In this simple situation the format is
FunctionName()
While the convention for variable identifiers is to start with a lowercase letter, the convention for function names is to
start with a capital letter. Hence HappyBirthdayEmily, not happyBirthdayEmily.
Can you predict what the program will do? Note the two function calls to HappyBirthdayEmily. To see, load
and run birthday2/birthday2.cs.
The execution sequence for the program is different from the textual sequence. Execution always starts in Main:
1. Line 13: Main is where execution starts, and initially proceeds sequentially.
2. Line 15: the function is called while this location is remembered.
3. Lines 5-11: Jump! The code of the function is executed for the first time, printing out the song.
4. End of line 15: Back from the function call; continue on.
5. Line 16: Just to mix things up, print out a “Hip, hip, hooray”.
6. Line 17: the function is called again while this location is remembered.
7. Lines 5-11: The function is executed again, printing out the song again.
8. End of line 17: Back from the function call, but at this point there is nothing more in Main, and execution stops.
Functions alter execution order in several ways: by statements not being executed as the definition is first read, and
then when the function is called during execution, jumping to the function code, and back at the the end of the function
execution.
Understanding the jumping around in the code with function calls is crucial. Be sure you follow the sequence detailed
above. In particular, be sure to distinguish function definition from function call.
If it also happens to be Andre’s birthday, we might define a function HappyBirthdayAndre, too. Think how to do
that before going on ....

3.2 Multiple Function Definitions

Here is example program birthday3/birthday3.cs where we add a function HappyBirthdayAndre, and call them
both. Guess what happens, and then load and try it:
Again, definitions are remembered and execution starts in Main. The order in which the function definitions are given
does not matter to C#. It is a human choice. For variety I show Main first. This means a human reading in order gets
an overview of what is happening by looking at Main, but does not know the details until reading the definitions of the
birthday functions.
Detailed order of execution:

42 Chapter 3. Defining Functions of your Own

 Introduction to Computer Science in C#, Release 2.0

1. Line 5: Start on Main

2. Line 7. This location is remembered as execution jumps to HappyBirthdayEmily
3. Lines 11-17 are executed and Emily is sung to.
4. Return to the end of Line 7: Back from HappyBirthdayEmily function call
5. Line 8: Now HappyBirthdayAndre is called as this location is remembered.
6. Lines 19-25: Sing to Andre
7. Return to the end of line 8: Back from HappyBirthdayAndre function call, done with Main; at the end of
the program
The calls to the birthday functions happen to be in the same order as their definitions, but that is arbitrary. If the two
lines of the body of Main were swapped, the order of operations would change, but if the order of the whole function
definitions were changed, it would make no difference in execution.
Functions that you write can also call other functions you write. In this case Main calls each of the birthday functions.

3.2.1 Poem Function Exercise

Write a program, poem.cs, that defines a function that prints a short poem or song verse. Give a meaningful name
to the function. Have the program call the function three times, so the poem or verse is repeated three times.

3.3 Function Parameters

As a young child, you probably heard Happy Birthday sung to a couple of people, and then you could sing to a new
person, say Maria, without needing to hear the whole special version with Maria’s name in it word for word. You had
the power of abstraction. With examples like the versions for Emily and Andre, you could figure out what change to
make it so the song could be sung to Maria!
Unfortunately, C# is not that smart. It needs explicit rules. If you needed to explain explicitly to someone how Happy
Birthday worked in general, rather than just by example, you might say something like this:
First you have to be given a person’s name. Then you sing the song with the person’s name inserted at the end of the
third line.
C# works something like that, but with its own syntax. The term “person’s name” serves as a stand-in for the actual
data that will be used, “Emily”, “Andre”, or “Maria”. This is just like the association with a variable name in C#.
“person’s name” is not a legal C# identifier, so we will use just person as this stand-in. It will be a variable in the
program, so it needs a type in C#. The names are strings, so the type of person is string.
In between the parentheses of the function definition heading, we insert the variable name person, preceded by its
type, string. Then in the body of the definition of the function, person is used in place of the real data for any
specific person’s name. Read and then run example program birthday4/birthday4.cs:
In the definition heading for HappyBirthday, person is referred to as a parameter, or a formal parameter. This
variable name is a placeholder for the real name of the person being sung to. In the definition we give instructions for
singing Happy Birthday without knowing the exact name of the person who might be sung to.
Main now has two calls to the same function, but between the parentheses, where there was the placeholder person
in the definition, now we have the actual people being sung to. The value between the parentheses here in the function
call is referred to as an argument or actual parameter of the function call. The argument supplies the actual data to
be used in the function execution. When the call is made, C# does this by associating the formal parameter name
person with the actual parameter data, as in an assignment statement. In the first call, this actual data is "Emily".
We say the actual parameter value is passed to the function for execution.

3.3. Function Parameters 43

Introduction to Computer Science in C#, Release 2.0

The execution in greater detail:

1. Lines 13: Execution starts in Main.
2. Line 15: Call to HappyBirthday, with actual parameter "Emily".
3. Line 5: "Emily" is passed to the function, so person = "Emily".
4. Lines 7-10: The song is printed, with "Emily" used as the value of person in line 9: printing
Happy Birthday, dear Emily.

5. End of line 15 after returning from the function call

6. Line 16: Call to HappyBirthday, this time with actual parameter "Andre"
7. Line 5: "Andre" is passed to the function, so person = "Andre".
8. Lines 7-10: The song is printed, with "Andre" used as the value of person in line 9: printing
Happy Birthday, dear Andre.

9. End of line 16 after returning from the function call, and the program is over.
The beauty of this system is that the same function definition can be used for a call with a different actual pa-
rameter variable, and then have a different effect. The value of the variable person is used in the third line of
HappyBirthday, to put in whatever actual parameter value was given.
This is the power of abstraction. It is one application of the most important principal in programming. Rather than
have a number of separately coded parts with only slight variations, see where it is appropriate to combine them using
a function whose parameters refer to the parts that are different in different situations. Then the code is written to be
simultaneously appropriate for the separate specific situations, with the substitutions of the right parameter values.

Note: Be sure you completely understand birthday4/birthday4.cs and the sequence of execution! It illustrates ex-
tremely important ideas that many people miss the first time! It is essential to understand the difference between
1. Defining a function (lines 5-11) with the heading including formal parameter name and type, where the code is
merely instructions to be remembered, not acted on immediately.
2. Calling a function with an actual parameter value to be substituted for the formal parameter, (with no type
included!) and have the function code actually run when the instruction containing the call is run. Also note that
the function can be called multiple times with different expressions as the actual parameter (line 15 and again in
line 16).

We can combine function parameters with user input, and have the program be able to print Happy Birthday for
anyone. Check out the Main method and run birthday_who/birthday_who.cs:
This last version illustrates several important ideas:
1. There are more than one way to get information into a function:
(a) Have a value passed in through a parameter (from line 18 to line 5).
(b) Prompt the user, and obtain data from the keyboard (lines 16-17).
2. It is a good idea to separate the internal processing of data from the external input from the user by the use of
distinct functions. Here the user interaction is in Main, and the data is manipulated in HappyBirthday.
3. In the first examples of actual parameters, we used literal values. In general an actual parameter can be an
expression. The expression is evaluated before it is passed in the function call. One of the simplest expressions
is a plain variable name, which is evaluated by replacing it with its associated value. Since it is only the value of
the actual parameter that is passed, not any variable name, there is no need to have a variable name used in an
actual parameter match a formal parameter name. (Here we have the value of userName in Main becoming
the value of person in HappyBirthday.)

44 Chapter 3. Defining Functions of your Own

 Introduction to Computer Science in C#, Release 2.0

3.3.1 Birthday Function Exercise

Make your own further change to birthday4/birthday4.cs and save it in your own project as birthday_many.cs:
Add a function call (but not another function definition), so Maria gets a verse, in addition to Emily and Andre. Also
print a blank line between verses. (There are two ways to handle the blank lines: You may either do this by adding a
print line to the function definition, or by adding a print line between all calls to the function. Recall that if you give
Console.WriteLine an empty parameter list, it just goes to the next line.)

3.4 Multiple Function Parameters

A function can have more than one parameter in a parameter list. The list entries are separated by commas. Each
formal parameter name is preceded by its type. For example the example program addition1/addition1.cs uses a
function, SumProblem, to make it easy to display many sum problems. Read and follow the code, and then run:
The actual parameters in the function call are evaluated left to right, and then these values are associated with the
formal parameter names in the function definition, also left to right. For example a function call with actual parameters,
F(actual1, actual2, actual3), calling a function F with definition heading:
static void F(int formal1, int formal2, int formal3)

acts approximately as if the first lines executed inside the called function F were
formal1 = actual1;
formal2 = actual2;
formal3 = actual3;

Functions provide extremely important functionality to programs, allowing tasks to be defined once and performed
repeatedly with different data. It is essential to see the difference between the formal parameters used to describe what
is done inside the function definition (like x and y in the definition of SumProblem) and the actual parameters (like 2
and 3 or 12345 and 53579) which substitute for the formal parameters when the function is actually executed. Main
uses three different sets of actual parameters in the three calls to SumProblem.

Warning: It is easy to confuse the heading in a function definition and a call to actually execute that function. Be
careful. In particular, do not list the types of parameters in a call’s actual parameter list. The actual parameters are
expressions involving terms that are already defined, not just being declared.

3.4.1 Quotient Function Exercise

Modify quotient.cs from Exercise for Quotients and save it as quotient_prob.cs. You should create a
function QuotientProblem with int parameters. Like in the earlier versions, it should print a full sentence with
inputs, quotient, and remainder. Main should test the QuotientProblem function on several sets of literal values,
and also test the function with input from the user.

3.5 Returned Function Values

You probably have used mathematical functions in algebra class, but they all had calculated values associated with
them. For instance if you defined
F(x)=x2

3.4. Multiple Function Parameters 45

Introduction to Computer Science in C#, Release 2.0

then it follows that F(3) is 32 , and F(3)+F(4) is 32 + 42

Function calls in expressions get replaced during evaluation by the value of the function.
The corresponding definition and examples in C# would be the following, taken from example program
return1.cs. Read and run:
The new C# syntax is the return statement, with the word return followed by an expression. Functions that return
values can be used in expressions, just like in math class. When an expression with a function call is evaluated, the
function call is effectively replaced temporarily by its returned value. Inside the C# function, the value to be returned
is given by the expression in the return statement.
Since the function returns data, and all data in C# is typed, there must be a type given for the value returned. Note
that the function heading does not start with static void. In place of void is int. The void in earlier function
headings meant nothing was returned. The int here means that a value is returned and its type is int.
After the function F finishes executing from inside
Console.WriteLine(F(3));

it is as if the statement temporarily became

Console.WriteLine(9);

and similarly when executing

Console.WriteLine(F(3) + F(4));

the interpreter first evaluates F(3) and effectively replaces the call by the returned result, 9, as if the statement tem-
porarily became
Console.WriteLine(9 + F(4));

and then the interpreter evaluates F(4) and effectively replaces the call by the returned result, 16, as if the statement
temporarily became
Console.WriteLine(9 + 16);

resulting finally in 25 being calculated and printed.

C# functions can return any type of data, not just numbers, and there can be any number of statements executed before
the return statement. Read, follow, and run the example program return2.cs:
Many have a hard time following the flow of execution with functions. Even more is involved when there are return
values. Make sure you completely follow the details of the execution:
1. Lines 12: Start at Main
2. Line 14: call the function, remembering where to return
3. Line 5: pass the parameters: firstName = "Benjamin"; lastName = "Franklin"
4. Line 7: Assign the variable separator the value ", "
5. Line 8: Assign the variable result the value of lastName + separator + firstName which is
"Franklin" + ", " + "Benjamin", which evaluates to "Franklin, Benjamin"
6. Line 9: Return "Franklin, Benjamin"
7. Line 14: Use the value returned from the function call so the line effectively becomes
Console.WriteLine("Franklin, Benjamin");, so print it.
8. Line 15: call the function with the new actual parameters, remembering where to return
9. Line 5: pass the parameters: firstName = "Andrew"; lastName = "Harrington"

46 Chapter 3. Defining Functions of your Own

 Introduction to Computer Science in C#, Release 2.0

10. Lines 7-9: ... calculate and return "Harrington, Andrew"

11. Line 15: Use the value returned by the function and print "Harrington, Andrew"
Compare return2/return2.cs and addition1/addition1.cs, from the previous section. Both use functions. Both print, but
where the printing is done differs. The function SumProblem prints directly inside the function and returns nothing
explicitly. On the other hand LastFirst does not print anything but returns a string. The caller gets to decide what
to do with the string, and above it is printed in Main.
In general functions should do a single thing. You can easily combine a sequence of functions, and you have more
flexibility in the combinations if each does just one unified thing. The function SumProblem in addition1/addition1.cs
does two thing: It creates a sentence, and prints it. If that is all you have, you are out of luck if you want to do something
different with the sentence string. A better way is to have a function that just creates the sentence, and returns it for
whatever further use you want. After returning that value, printing is one possibility, done in addition2/addition2.cs:
In class recommendation: Improve example painting/painting.cs with functions. Copy it to a file
painting_input.cs in your own project and modify it.

3.5.1 Quotient String Return Exercise

Create quotient_return.cs by modifying quotient_prob.cs in Quotient Function Exercise so that the

program accomplishes the same thing, but everywhere:
• Change the QuotientProblem function into one called QuotientString that merely returns the string rather
than printing the string directly.
• Have Main print the result of each call to the QuotientString function.

3.6 Two Roles: Writer and Consumer of Functions

The remainder of this section covers finer points about functions that you might skip on a first reading.
We are only doing tiny examples so far to get the basic idea of functions. In much larger programs, functions are
useful to manage complexity, splitting things up into logically related, modest sized pieces. Programmers are both
writers of functions and consumers of the other functions called inside their functions. It is useful to keep those two
roles separate:
The user of an already written function needs to know:
1. the name of the function
2. the order and meaning of parameters
3. what is returned or produced by the function
How this is accomplished is not relevant at this point. For instance, you use the work of the C# development team,
calling functions that are built into the language. You need know the three facts about the functions you call. You do
not need to know exactly how the function accomplishes its purpose.
On the other hand when you write a function you need to figure out exactly how to accomplish your goal, name
relevant variables, and write your code, which brings us to the next section.
The jargon for these parts are the interface (for the consumer) and the implementation (for the programmer, who must
be sure to satisfy the public interface).

3.6. Two Roles: Writer and Consumer of Functions 47

Introduction to Computer Science in C#, Release 2.0

3.7 Local Scope

For the logic of writing functions, it is important that the writer of a function knows the names of variables inside
the function. On the other hand, if you are only using a function, maybe written by someone unknown to you, you
should not care what names are given to values used internally in the implementation of the function you are calling.
C# enforces this idea with local scope rules: Variable names initialized and used inside one function are invisible
to other functions. Such variables are called local variables. For example, an elaboration of the earlier program
return2/return2.cs might have its lastFirst function with its local variable separator, but it might also have
another function that defines a separator variable, maybe with a different value like "\n". They would not
conflict. They would be independent. This avoids lots of errors!
For example, the following code in the example program bad_scope/bad_scope.cs causes a compilation error. Read it
and try to run it, and see:
The compilation error that Mono gives is pretty clear:
The name ‘x’ does not exist in the current context.
The context for x is the function F, not Main. We will fix this error below.
If you do want local data from one function to go to another, define the called function so it includes parameters! Read
and compare and try the program good_scope/good_scope.cs:
With parameter passing, the parameter name x in the function F does not need to match the name of the actual
parameter in the calling function Main. The definition of F could just as well have been:
static void F(int whatever)
{
Console.WriteLine(whatever);
}

3.8 Static Variables

You may define static variables (variables defined inside the class, but outside of any function definition). These
variables are visible inside all of your functions. Instead of local scope, static variables have class scope. It is good
programming practice generally to avoid defining static variables and instead to put your variables inside functions and
explicitly pass them as parameters where needed. One common exception will arise when we get to defining objects.
For now a good reason for static variables is constants: A constant is a name that you give a fixed data value to. You
can then use the name of the fixed data value in expressions anywhere in the class. A simple example program is
constant/constant.cs:
See that PI is used in two functions without being declared locally.
By convention, names for constants are all capital letters.

3.9 Not using Return Values

Some functions return a value, and get used as an expression in a larger calling statement. The calling statement uses
the value returned. Usually the only effect of such a function is in the value returned.
Some functions are void, and get used as a whole instruction in your code: Without returning a value, the only way to
be useful is to do something that leaves some lasting side effect: make some change to the system that persists after the
termination of the function and its local variables disappear. The only such effect that we have seen so far is to print
something that remains on the console screen. Later we will talk about other persistent changes to values in objects,
locations in files, ....

48 Chapter 3. Defining Functions of your Own

 Introduction to Computer Science in C#, Release 2.0

Usually there is this division in the behavior of functions, returning a value or not:
1. void: do something as whole instruction, with a side effect in the larger system.
2. Return a value to use in a larger calling statement
It is legal to do both: accomplish something with a side effect in the system, and return a value, and sometimes use
the function only for its side effect. We will see examples of that later, like in Sets.
This later advanced use will mean that the compiler needs to permit the programmer to ignore a returned value, and
use a function returning a value as a whole statement. However, this means that the compiler cannot catch a common
logical error, forgetting to immediately use a returned value that your program logic really needs. For example with
this definition:
static int CalcResult(int param)
{
int result;
// ....
result = ....;
return result;
}

you might try to use CalcResult in this bad code, intending to use the result from CalcResult:
static void BadUseResult(int x)
{
int result = 0;
CalcResult(x);
Console.WriteLine(result);
}

In fact you would always print 0, ignoring the result calculated in CalcResult. The reason is the scope rules for
functions: The local variable result disappears when the CalcResult function returns. It has no meaning later
in the calling function, BadUseResult.
We set up the worst situation, where there is a logical error, but not an error shown by the compiler. More commonly a
student leaves out the int result = 0; line, incorrectly relying on the declaration of result in CalcResult.
At least in that situation a compiler error brings attention to the problem.
You can actually use the result from CalcResult with either
int value = CalcResult(x); //store the returned value in an assignment!
Console.WriteLine(value);

or
Console.WriteLine(CalcResult(x)); // immediately use the returned value

This version works as long as you do not need the returned value in another place, later, since you do not remember it
past that one statement.

3.10 Library Classes

In Returned Function Values, the suggestion was made to look at the Painter class and split out repeated ideas into
functions, leading to a function to prompt the user and return a double value. The same section included the example
program addition2/addition2.cs. In that case there were repeated prompts for integers. Clearly another common
situation is to prompt for a string. We can create functions to do all these things and more, and embed them into a
class for an interactive program.

3.10. Library Classes 49

Introduction to Computer Science in C#, Release 2.0

A neater thing is to put them as a class in a separate library that can be used directly for multiple programs. We can
create functions PromptLine, PromptInt, and PromptDouble, and put them in their own class, UIF (for User
Input First version) in project ui’s file uif.cs. We explain the namespace line after the code:
We have been using System in every program. System is a namespace that collects a particular group of class
names, making them available to the program, and distinguishes them form any classes in a different namespace that
might have the same class names.
Once we start writing and using multiple classes at once, it is a good idea for us to specify our own namespace. We
will consistently use IntroCS in our multi-file examples in this book.
Specifying a namespace makes it possible for all other classes in the same namespace to reference public parts of the
current class, and vice-versa.
Public classes and functions start their heading with public.
The code included in a namespace is enclosed in braces, so the general syntax is
namespace name
{
class definition(s)...
}
We will keep user input library classes like this one, uif.cs, in a project ui.
Notice that the functions we want accessible in UIF are all marked public, so that any class can use them.
We can write a modified example addition program, addition3/addition3.cs, as an example of using UIF:
To allow access to UIF, we have added the IntroCS namespace for the class. To reference the static functions in the
different class UIF, we put UIF. (with the dot) at the start of each reference to a static function in the class UIF.

Warning: In Xamarin Studio, if you use a file from a library project, be sure that the current project includes a
reference to the library project. If you expand the references in the Xamarin Studio project addition3, by clicking
on the References line in the solution pad, you should see the project ui.

Shortly we will come to making your own Library Projects in Xamarin Studio.
Though we have not discussed all the C# syntax needed yet, there is also an improved class UI in the ui project that we
discuss later. It includes all the function names in UIF, and keeps your program from bombing out if the user enters
an illegal format for a number.

3.10.1 Function Documentation

In keeping with Two Roles: Writer and Consumer of Functions, in future you will be a consumer of the library classes.
It is particularly important to document library classes with the interface information users will need. Documentation
could be written in a separate document, but much developer history has shown that such documentation does not tend
to either get written in the first place, or not updated well to stay consistent with updates in the code. Inconsistent
documentation is useless. Documentation is much more likely to be seen and maintained by the implementers if it sits
right with the code, like our comments before the class and function headings.
You will note that instead of the usual line comment syntax //, we have added an extra /, making ///. That will
also start a comment. (The third / is technically just a part of the comment.) There is a special reason for the
notation: Though it is convenient for the implementer of code to have the documentation right with the code, a user
of the functions only needs the interface information found in good documentation. The /// lines before heading are
specially recognized by separate automatic documentation generating programs.
There are many documentation generating programs and conventions. For now we will just use plain text in the ///
lines. This is recognized by the Xamarin Studio system. If you open our examples solution, in Xamarin Studio,

50 Chapter 3. Defining Functions of your Own

 Introduction to Computer Science in C#, Release 2.0

and edit window for addition3/addition3.cs, you can place your mouse over UIF and a popup window shows the UIF
class heading documentation.
If you move the mouse over PromptInt, you should see the popup label showing the function signature and the
function documentation. If you change the two /// lines in uif.cs above the PromptInt heading to start with just
//. you should no longer be able to see the documentation part of the popup for PromptInt in the addition3.cs
edit window. (Be sure to change back to ///.)
There are more elaborate documentation conventions that can be used for Xamarin Studio and other documentation
generation programs. We defer that discussion to a later appendix.
This documentation also works inside a single program file. If you have a long program with lots of functions defined,
this can also be helpful when calling one of your own functions. You can avoid jumping around to be reminded of the
signature and use of your functions.

3.10.2 Library Projects in Xamarin Studio

Try adding a reference yourself. Follow these instructions:

1. In your own Xamarin Studio solution, start to add a project, but instead of leaving Console Project selected in
the dialog window, select Library Project.
2. Then add the project name ui, and continue like when starting previous projects.
3. Copy in the .cs files from our ui project, uif.cs and ui.cs. Now you have your library project.
4. Create another regular Console project, addition3, in your same solution, and copy in our addition3/addition3.cs,
so that is the only file.
5. In the Solutions pad, in your addition3 project, click on the References entry just inside the project. You should
see that the project is automatically set up to reference System.
6. Open the local menu for the References, and select Edit References.
7. Click the Projects tab in the window that pops up. This limits the length of the list that you search.
8. Possibly after scrolling down, find the recently made ui project and check the box beside it.
9. Click OK in the bottom right corner of the window. Now look at the References again. You should see ui listed!
10. Run your addition3 project.
You only need to add a library project once, but every further project that needs it, must have a reference to the library
project added. You might try another for yourself with the next exercise!

3.10.3 Quotient UI Exercise

Create quotient_u_i.cs by modifying quotient_return.cs in Quotient String Return Exercise so that the
program accomplishes the same thing, but use the UIF class for all user input.

3.11 Static Function Summary

This chapter has introduced static functions: those used in procedural programming as opposed to Instance Methods
used to implement object-oriented programming.
References in square brackets link to fully discussions of summary items below.

3.11. Static Function Summary 51

Introduction to Computer Science in C#, Release 2.0

3.11.1 Function definition

1. The general syntax for defining a static function is

static returnTypeOrVoid FunctionName ( formal parameter list )
{
statements in the function body...
}
2. The formal parameter list can be empty or contain one or more comma separated formal parameter entries.
[Function Parameters] Each formal parameter entry has the form
type parameterName
3. If the function is going to be called from outside its class, the heading needs to start with public before the
static. [Library Classes]
4. If returnTypeOrVoid in the heading is not void, there must be a return statement in the function body. A
return statement has the form
return expression ;
where the expression should be of the same type as in returnTypeOrVoid. Execution of the function terminates
immediately when a return statement is reached. [Returned Function Values]
5. Execution of a program starts at a function with a heading including
static void Main
Thus far we have only discussed having an empty parameter list in the heading of the definition of Main, and
we defer discussion of Parameters to Main until we have introduced One Dimensional Arrays.
6. There are various conventions for putting documentation just above the headings of function definitions. The
official format, specified by C# and recognized by Xamarin Studio, involves putting the function interface
description on consecutive lines starting with ///. [Function Documentation]

3.11.2 Function Calls

1. A function call takes the form

FunctionName ( actual parameter list )
A function call makes the function definition be executed.
2. The actual parameter list is a comma separated list of the same length as the formal parameter list. Each entry
is an expression. The entries in an actual parameter list do not include type declarations.
Effectively, the function execution starts by assigning to each formal parameter variable the corresponding
value from evaluating the actual parameter expression. In particular, that means the actual parameter values
must be allowed in an assignment statement for a variable of the formal parameter’s type! [Multiple Function
Parameters]
3. If the function has return type void, it can only be used syntactically as an entire statement (with a semicolon
added). After the function call completes, execution continues with the next statement.
4. If there is a non-void return type, then the function call is syntactically an expression in the statement where
is appears. The execution of such a function must reach a return statement. The value of the function-call
expression is the value of the expression in this return statement. [Returned Function Values]
5. A function with a return value can also legally be used as a whole statement. In this case the return value is lost.
Though legal, this is often an error! [Not using Return Values]

52 Chapter 3. Defining Functions of your Own

 Introduction to Computer Science in C#, Release 2.0

3.11.3 Scope

1. A variable declared inside a function definition is called a local variable. This declaration may be in either the
formal parameter list or in the body of the function. [Local Scope]
2. A local variable comes into existence after the function is called, and ceases to exist after that function call
terminates. A local variable is invisible to the rest of the program. Its scope is just within that function. Its
lifetime is just through a single function call. Its value may be transferred outside of the function scope by
standard means, principally:
• If it is the expression in a return statement, its value is sent back to the caller.
• It can be passed as an actual parameter to a further function called within its scope.
[Local Scope]

3.11.4 Static Variables

1. There may be a declaration prefaced by the word static that appears inside a class and outside of any function
definition in the class. Static variable are visible within the functions of the class, and may be used by the
functions. [Static Variables]
2. A common use of a static variable is to give a name to a constant value used in multiple functions in the class.
[Static Variables]

3.11. Static Function Summary 53

Introduction to Computer Science in C#, Release 2.0

54 Chapter 3. Defining Functions of your Own

 CHAPTER

FOUR

BASIC STRING OPERATIONS

4.1 String Indexing

Strings are composed of characters. In literals be careful of the different kinds of quotes: single for individual char-
acters for type char and double for strings of 0 or more characters. For example, ‘u’ (single quotes) is a char type
literal, while “u” is a string literal, referencing a string object. While “you” is a legal string literal, ‘you’ generates a
compiler error (too many characters - only one allowed).
Many of the operations on strings depend upon referring to the positions of characters in the string. A position is given
by a numerical index number. In C#, positions are counted starting at 0, not 1. The indices of the characters in the
string “coding” are labeled:
Index 0 1 2 3 4 5
Character c o d i n g
There are 6 characters in "coding", while the last index is 5.

Warning: Because the indices start at 0, not 1, the index of the last character is one less that the length of the
string. This is a common source of errors.

You can easily create an expression that refers to an individual character inside a string. Use square braces around the
index of the character:
csharp> string s = "coding";
csharp> s[2];
’d’
csharp> s[0];
’c’
csharp> s[5];
’g’
csharp> string greeting = "Bonjour";
csharp> greeting[1];
’o’

Note from the single quotes that the result is a char in each case.
C# does not allow the typography for normal mathematical subscripts, like 𝑠2 . There is a correspondence with index
notation, so s[2] is sometimes spoken as “s sub 2”. The indices are sometimes referred to as subscripts.

4.2 Some Instance Methods and the Length Property

Strings are a special type in C#. We have used string literals as parameters to functions and we have used the special
concatenation operator +. Thus far we have not emphasized the use of objects, or even noted what is an object. In

55
Introduction to Computer Science in C#, Release 2.0

fact strings are objects. Like other objects, strings have a general notation for functions that are specially tied to the
particular type of object. These functions are called instance methods. They always act on an object of the particular
class, but a reference to the object is not placed inside the parameter list, but before the method name and a dot as in:
csharp> string s = "hello";
csharp> s.ToUpper();
"HELLO"

ToUpper (converting to upper case) is particular action that makes sense with strings. It take s (the string object
reference before the dot in this example) and returns a new string in upper case, based on s. Since this action depends
only on the string itself, no further parameters are necessary, and the parentheses after the method name are empty.
The general method syntax is
object-reference.methodName (further-parameters )
More string methods are listed below, some with further parameters.
Data can also be associated with object properties. A property of a string is its Length (an int). References to
property values use dot notation but do not have a parameter list in parentheses at the end:
csharp> string s = "Hello";
csharp> s.Length;
5
csharp> "".Length;
0

Be careful: Though 5 is the length of s in the example above, the last character in s is s[4]. Using s[5] would
generate an IndexOutOfRangeException.
String objects have associated string methods which can be used to manipulate string values. There are an enormous
number of string methods, but here are just a few of the most common ones to get you started. The string object
to which the method is being applied is referred to as this string in the descriptions. After the methods, the length
property is also listed. In the heading this object is not shown explicitly, so be careful when applying these methods
and the length property: In actual use in your programs they must be preceded by a reference to a string, followed by
a dot, as shown in all the examples. The reference to this string can be a variable name, a literal, or any expression
evaluating to a string.

4.2.1 Summary of String Length and Some Instance Methods

int IndexOf(string target) Returns the index of the beginning of the first occurrence of the string
target in this string object. Returns -1 if target not found. Examples:
csharp> string greeting = "Bonjour", part = "jo";
csharp> greeting.IndexOf(part);
3
csharp> greeting.IndexOf("jot");
-1

string Substring(int start) Returns the substring of this string object starting from index start
through to the end of the string object. Example:
csharp> string name = "Sheryl Crow";
csharp> name.Substring(7);
"Crow"

string Substring(int start, int len) Returns the substring of this string object starting from index
start, including a total of len characters. Example:

56 Chapter 4. Basic String Operations

 Introduction to Computer Science in C#, Release 2.0

csharp> string name = "Sheryl Crow";

csharp> name.Substring(3,5);
"ryl C"

string ToUpper() Return a string like this string, except all in upper case. Example:
csharp> "Hi Jane!".ToUpper();
"HI JANE!"

string ToLower() Return a string like this string, except all in lower case. Example:
csharp> "Hi Jane!".ToLower();
"hi jane!"

int Length Property referring to the length of this string object. Example:
csharp> string greeting = "Bonjour";
csharp> greeting.Length; //no parentheses
7

All of these methods that return a string return a new string. No string method alters the original string. Strings are
immutable: They are objects that cannot be changed after they are first produced.
Further string methods are introduced in More String Methods.
Time to reflect, thinking back to Learning to Solve Problems. Without forcing all the code details on yourself, how
can you concisely say what powers you have with strings so far? Remember that kernel.
With strings you can: Index characters, find a part; extract a part; convert case; determine length. These may not be
evocative phrases for you. Find your own.

4.3 A Creative Problem Solution

Thus far the exercises and examples suggested have been of a very simple form, where the idea of the steps should
have been pretty clear, and the main issue was just translating syntax into C#, one instruction at a time.
We still have a lot of syntax to concentrate on, but still, early on, we wanted to get in some real thought of problem
solving. To get very interesting you need a number of options that might be combined in a variety of ways. The short
list of string methods just introduced is likely give us enough to think about....
Here is a basic string manipulation problem: given a string, like, "It was the best of times.", find and
replace a specified part of it by another string. For instance replace "best" by "worst". In this example we would
get the result: "It was the worst of times.".
It is very important to give concrete examples to illustrate the idea desired. Our human brains may be very quick to
see a solution like this in a very concrete case, but what about making it general?
First this seems like a basic logical operation worthy of a function or method, so we need a heading. (Confession:
there are methods in the class string for replacement, but this is a good learning exercise, so we are starting over on
our own.) Since we cannot change the string class, we will write a static function to generate the new string.
For simplicity at the moment we will only change the first occurrence, and for now we will assume the replacement
makes sense. The following heading (with documentation) should work:
As soon as we have the calling interface, it is good to be thinking of the tests it should pass. Here is a Main program
written to test the function in different ways and display the results:
Writing tests first is a good idea to focus you on what really needs to be accomplished, and then running tests later is
a snap!

4.3. A Creative Problem Solution 57

Introduction to Computer Science in C#, Release 2.0

The human brain and eyes are fabulous in the way they process many things in parallel and use tools you have
accumulated over a lifetime. In particular this substitution idea should seem pretty reasonable, and given any specific
concrete example, you are likely to be able to solve it instantly, with very little conscious effort. Once it becomes
a programming problem, with parameters stated in general, with just placeholder names like s and target, and
given the limited set of approaches you have in a programming language, the complexion of this problem changes
completely. Many students guess the general problem will be nearly as simple as the concrete examples they do in
their heads, and then get very discouraged when the answer does not flow out of them. In fact it takes practice and
experience, and it is easier to handle if you acknowledge that up front!
So let’s start in with the practice, and gain some experience. With s, target, and replacement all being general,
this problem could easily be too much to contemplate at once, so let us replace concrete examples by generality
gradually. The idea is to get to the end. Rather than trying to jump a chasm, we can take small steps and go around.
A basic idea is to make small incremental changes, test at each stage, and gradually see more of the tests (that you
have already written) be satisfied. Also, if you make a mistake and screw up something that worked before, you can
generally focus on the small addition to see where the mistakes were. 1
This also avoids you needing to keep too much in your head at once.
We do have code written already: The test code. Start by writing something that will trivially satisfy the first concrete
test. The body of the function can be just:
return "It was the worst of times";

This is a tiny, easy, silly looking step, but it does accomplish two things: It makes sure we can produce output in the
proper string form, and the test code runs, passing the first test.
Now we gradually get more complicated. We will continue to assume target and replacement are as in the
original example, and target is in the same place in s, but suppose we imagine each of the other characters in s
may be something different:
"???????????best??????????"

Now we have to start thinking about what we have to work with. We have a string, and we have string methods. Have
a look at the ideas of each method (exact syntax not important at the moment). Clearly we are going to have to deal
with parts of strings, and the methods to deal with parts involve indices, so let us add to our visual model:
Index: 0123456789012345678901234
s: ???????????best??????????

Continue in class.... The example program stub is string_manip_stub/string_manip.cs.

In general, when given a project with “stub” in it, you should copy the files into a project of your own and make
modifications. Though the original version should compile and run, it does not do much without your additions. In
stubs where you need to complete a function with a return value, you will often see a dummy choice for the return
statement, just so the stub compiles. Where the return type is string "Not implemented" is a handy temporary
choice.
When you have that function version, test it. You will need to rename our incremental variations so the current version
has the name used in Main.
What might further advances toward full generality be, in small steps? We pinned best at a specific location. We
could remove that assumption. The location will still be important, but we do not know it ahead of time....
A further advance would be a version that is complete in all ways, except we still assume target is in s, but beyond
that, do not assume what the three parameters are.
Finally we should allow s to not contain target.
1 We will not go far into the history of software engineering practice here, but these incremental problem solving methods were first widely

introduced as a part of extreme programming. That name gives you an idea of the newness at the time.

58 Chapter 4. Basic String Operations

 Introduction to Computer Science in C#, Release 2.0

The testing regime in Main is clear to understand and write, but pretty primitive. You have to look at a lot of output
every time you test. We will come up with better testing schemes later.

4.4 Lab: String Operations

4.4.1 Goals for this lab:

1. Explore some of the properties of the pre-defined String class.

2. Write conditional statements.
3. Think about problem solving.
This lab depends on the introductory material in earlier in this chapter, particularly keep handy Summary of String
Length and Some Instance Methods. Be mindful of the processes developed in class filling in A Creative Problem
Solution.
Parts 2 and 4 also depend on Decisions through More Conditional Expressions.
Design, compile and run a single C# program to accomplish all of the following tasks. Add one part at a time and test
before trying the next one. The program can just include a Main method, or it is neater to split things into separate
methods (all static void, with names like ShowLength, SentenceType, LastFirst1, LastFirst), and have Main call all
the ones you have written so far (or for testing purposes, just the one you are working on, with the other function calls
commented out). Practice using a library reference: Use the UIF class for user input.
If using Xamarin Studio, create a new project in a solution in which you already have added the ui library project. Add
the ui project as a reference for the lab project. Make sure your program has namespace IntroCS; to match the
ui project. Beware of modifying the sample program generated by Xamarin Studio - it will use the project name for a
namespace. We are never going to use that.
1. Read a string from the keyboard and print the length of the string, with a label.
2. Read a sentence (string) from a line of input, and print whether it represents a declarative sentence (i.e. ending
in a period), interrogatory sentence (ending in a question mark), or an exclamation (ending in exclamation point)
or is not a sentence (anything else).
This may be the first time you write a conditional statement. It makes sense to only make small changes at
once and build up to final code. First you might just code it to check if a sentence is declarative or not. Then
remember you can test further cases with else if (...).
3. Read a name from a line of input. Assume first and last names are separated by a space. Print last name first
followed by a comma and a space, followed by the first name. For example, if the input is "Marcel Proust",
the output is "Proust, Marcel".
4. Improve the previous part, so it also allows a single name without spaces, like “Socrates”, and prints the original
without change. If there are two parts of the name, it should work as in the original version.
Run the program (with parts 1, 2 and 4 active) from a terminal window and show your TA when you are done. You
should run it twice to show off both paths through part 4. Alternately have the main program just call part 4 twice.

4.4. Lab: String Operations 59

Introduction to Computer Science in C#, Release 2.0

60 Chapter 4. Basic String Operations

 CHAPTER

FIVE

DECISIONS

5.1 Conditions I

Thus far, within a given function, instructions have been executed sequentially, in the same order as written. Of course
that is often appropriate! On the other hands if you are planning out instructions, you can get to a place where you say,
“Hm, that depends....”, and a choice must be made. The simplest choices are two-way: do one thing is a condition is
true, and another (possibly nothing) if the condition is not true.
More syntax for conditions will be introduced later, but for now consider simple arithmetic comparisons that directly
translate from math into C#. Try each line separately in csharp
2 < 5;
3 > 7;
int x = 11;
x > 10;
2 * x < x;

You see that conditions are either true or false (with no quotes!). These are the only possible Boolean values
(named after 19th century mathematician George Boole). You can also use the abbreviation for the type bool. It is
the type of the results of true-false conditions or tests.
The simplest place to use conditions is in a decision made with an if statement.
We will consider More Conditional Expressions later, but this is a quick start with the easiest ones.

5.2 Simple if Statements

Run the example program, suitcase/suitcase.cs. Try it at least twice, with inputs: 30 and then 55. As you an see, you
get an extra result, depending on the input. The main code is:
The lines labeled 3-5 are an if statement. It reads pretty much like English. If it is true that the weight is greater
than 50, then print the statement about an extra charge. If it is not true that the weight is greater than 50, then skip the
part right after the condition about printing the extra luggage charge. In any event, when you have finished with the
if statement (whether it actually does anything or not), go on to the next statement. In this case that is the statement
printing “Thank you”. An if statement only breaks the normal sequential order inside the if statement itself.
The general C# syntax for a simple if statement is
if ( condition )
statement
Often you want multiple statements executed when the condition is true. We have used braces before. We have not
said what they do technically, syntactically: braces around a group of statements technically makes a single compound
statement. So the pattern commonly written is:

61
Introduction to Computer Science in C#, Release 2.0

if ( condition ) {
one or more statements
}
If the condition is true, then do the statement(s) in braces. If the condition is not true, then skip the statements in
braces. The indentation pattern is also illustrated. Recall the compiler does not care about the amount of whitespace,
but humans do. In general indent the statements inside a compound statement. We will see later that there is good
reason to use this format with braces even if there is just one statement inside the braces.
Another fragment as an example:
if (balance < 0) {
transfer = -balance;
// transfer enough from the backup account:
backupAccount = backupAccount - transfer;
balance = balance + transfer;
}

The assumption in the example above is that if an account goes negative, it is brought back to 0 by transferring money
from a backup account in several steps.
In the examples above the choice is between doing something (if the condition is true) or nothing (if the condition
is false). Often there is a choice of two possibilities, only one of which will be done, depending on the truth of a
condition....

5.3 if-else Statements

Run the example program, clothes/clothes.cs. Try it at least twice, with inputs 50 and then 80. As you can see, you
get different results, depending on the input. The main code of clothes/clothes.cs is:
The lines labeled 2-7 are an if-else statement. Again it is close to English, though you might say “otherwise”
instead of “else” (but else is shorter!). There are two indented statements in braces: One, like in the simple if
statement, comes right after the if condition and is executed when the condition is true. In the if-else form this
is followed by an else (lined up under the if by convention), followed by another indented statement enclosed in
braces that is only executed when the original condition is false. In an if-else statement exactly one of two possible
parts in braces is executed.
A final line is also shown that is not indented, about getting exercise. The if and else clauses each only embed a single
(possibly compound) statement as option, so the last statement is not part of the if-else statement. Instead it is a
part of the normal sequential flow of statements. It is always executed after the if-else statement, no matter what
happens inside the if-else statement. Again: inside the if-else there is a choice made, but the whole if-else
construction is a single larger statement, which exists in the normal sequential flow. The compiler does not require the
indentation of the if-true-statement and the if-false-statement, but it is a standard style convention.
The general C# if-else syntax is
if ( condition ) {
statement(s) for if-true
}
else {
statement(s) for if-false
}
The statements chosen based on the condition can be any kind of statement. This is the suggested form, but as with
the plain if statement, the if-true compound statement or the if-false compound statement can be replace by a single
statement without braces, except in one otherwise ambiguous situation discussed later with two ifs and an else.

62 Chapter 5. Decisions
 Introduction to Computer Science in C#, Release 2.0

5.3.1 Scope With Compound Statements

The section Local Scope referred to function bodies, which happen to be enclosed in braces, making the function body
a compound statement. In fact variables declared inside any compound statement have their scope restricted in inside
that compound statement.
As a result the following code makes no sense:
static int BadBlockScope(int x)
{
if ( x < 100) {
int val = x + 2;
}
else {
int val = x - 2:
}
return val;
}

The if-else statement is legal, but useless, because whichever compound statement gets executed, val ceases being
defined after the closing brace of its compound statement, so the val in the return statement has not been declared or
given a value. The code would generate a compiler error.
If we want val be used inside the braces and to make sense past the end of the compound statement, it cannot be
declared inside the braces. Instead it must be declared before the compound statements that are parts of the if-else
statement. A local variable in a function declared before a nested compound statement is still visible (in scope) inside
that compound statement. The following would work:
There is even more subtlety here than meets the eye: An if-else statement can generally be rewritten as two simple
if statements (though it is less efficient and less clear). The two if statements would use opposite conditions, as in
this variation:
Notie that in this variation we added an initialization for val to be 0, though the value of the initialization is never
used: val is guaranteed to be assigned a value in one of the if statements before its value is used in the return
statement.
Open Xamarin Studio with the examples solution, and open ok_if_scope/‘ok_if_scope.cs in the edit window. The
last function, OkScope2, is the one shown above. Now remove the logically unnecessary = 0 initialization for val
so the line is just int val;. As the comment says, an error should appear (at least after you try to compile the
program). The error will say that there is an uninitialized local variable! Why?
For safety the C# compiler has some basic analysis to check that every local variable gets given a value before its value
is used. In the OkScope function there is no one place where val gets an initial value, but the compiler is smart
enough to see that one of the branches of any if-else statement is always taken, and val gets a value in each, so there
is no problem.
The compiler analysis is not complete: It does not actually evaluate any expressions. This is good enough to catch
many initialization errors that coders make, but it is not sufficient in general: We can see this from the altered
OkScope2.
The original code shows the fix: Give a dummy initialization that is never used in execution, but keeps the compiler
happy.
Although this extra initialization is annoying, the extra step is rarely needed. Meanwhile it is very easy to forget to
give a value to a local variable before use! Having the error caught quickly by the compiler is very handy, offsetting
the extra work when the compiler gives this error unnecessarily.

5.3. if-else Statements 63

Introduction to Computer Science in C#, Release 2.0

5.4 More Conditional Expressions

All the usual arithmetic comparisons may be made in C#, but many do not use standard mathematical symbolism,
mostly for lack of proper keys on a standard keyboard. (If you are looking at the following table in the web version,
you may need an up-to-date browser to see the mathematical symbols correctly, and also the mathematical expressions
later in the text.)
Meaning Math Symbol C# Symbols
Less than < <
Greater than > >
Less than or equal ≤ <=
Greater than or equal ≥ >=
Equals = ==
Not equal ̸ = !=
There should not be space between the two-symbol C# substitutes.
Notice that the obvious choice for equals, a single equal sign, is not used to check for equality. An annoying second
equal sign is required. This is because the single equal sign is already used for assignment in C#, so it is not available
for tests.

Warning: It is a common error to use only one equal sign when you mean to test for equality, and not make an
assignment!

Tests for equality do not make an assignment. Tests for equality can have an arbitrary expression on the left, not just a
variable.
All these tests work for numbers, and characters. Strings can also be compared, most often for equality (==) or
inequality (!=), though they also have a defined order, so you can use <, for instance.
Predict the results and try each line in csharp:
int x = 5;
x;
x == 5;
x == 6;
x;
x != 6;
x = 6;
6 == x;
6 != x;
"hi" == "h" + "i";
"HI" != "hi";
string s = "Hello";
string t = "HELLO";
s == t;
s.ToUpper() == t;

An equality check does not make an assignment. Strings equality tests are case sensitive.
Try this: Following up on the discussion of the inexactness of float arithmetic, confirm that C# does not consider .1 +
.2 to be equal to .3: Write a simple condition into csharp to test.

Pay with Overtime Example

Given a person’s work hours for the week and regular hourly wage, calculate the total pay for the week, taking into
account overtime. Hours worked over 40 are overtime, paid at 1.5 times the normal rate. This is a natural place for a

64 Chapter 5. Decisions
 Introduction to Computer Science in C#, Release 2.0

function enclosing the calculation.

Read the setup for the function:
The problem clearly indicates two cases: when no more than 40 hours are worked or when more than 40 hours are
worked. In case more than 40 hours are worked, it is convenient to introduce a variable overtimeHours. You are
encouraged to think about a solution before going on and examining mine.
You can try running my complete example program, wages1/wages1.cs, also shown below.
This program also introduces new notation for displaying decimal numbers:
In the format string are {1:F2} and {2:F2}: Inside the braces, after the parameter index, you see a new part, :F2.
The part after the colon gives optional formatting information. In this case display with the decimal point fixed (hence
the f) so 2 places beyond the decimal point are shown. Also the result is rounded. This is appropriate for money
with dollars and cents. Replace the 2 to display a different number of digits after the decimal point. More formatting
instructions will be discussed later.
Below is an equivalent alternative version of the body of CalcWeeklyWages, used in wages2/wages2.cs. It uses
just one general calculation formula and sets the parameters for the formula in the if statement. There are generally
a number of ways you might solve the same problem!

5.4.1 Graduate Exercise

Write a program, graduate.cs, that prompts students for how many credits they have. Print whether of not they
have enough credits for graduation. (At Loyola University Chicago 120 credits are needed for graduation.)

5.5 Multiple Tests and if-else Statements

Often you want to distinguish between more than two distinct cases, but conditions only have two possible results,
true or false, so the only direct choice is between two options. As anyone who has played “20 Questions” knows,
you can distinguish more cases with further questions. If there are more than two choices, a single test may only reduce
the possibilities, but further tests can reduce the possibilities further and further. Since most any kind of statement can
be placed in the sub-statements in an if-else statement, one choice is a further if or if-else statement. For
instance consider a function to convert a numerical grade to a letter grade, ‘A’, ‘B’, ‘C’, ‘D’ or ‘F’, where the cutoffs
for ‘A’, ‘B’, ‘C’, and ‘D’ are 90, 80, 70, and 60 respectively. One way to write the function would be to test for one
grade at a time, and resolve all the remaining possibilities inside the next else clause. If we do this consistent with
our indentation conventions so far:
static char letterGrade(double score)
{
char letter;
if (score >= 90) {
letter = ’A’;
}
else { // grade must be B, C, D or F
if (score >= 80) {
letter = ’B’;
}
else { // grade must be C, D or F
if (score >= 70) {
letter = ’C’;
}
else { // grade must D or F
if (score >= 60) {
letter = ’D’;

5.5. Multiple Tests and if-else Statements 65

Introduction to Computer Science in C#, Release 2.0

}
else {
letter = ’F’;
}
} //end else D or F
} // end of else C, D, or F
} // end of else B, C, D or F
return letter;
}

This repeatedly increasing indentation with an if statement in the else clause can be annoying and distracting. Here
is a preferred alternative in this situation, that avoids all this further indentation: Combine each else and following
if onto the same line, and note that the if part after each else is just a single (possibly very complicated) statement.
This allows the elimination of some of the braces:
A program testing the letterGrade function is in example program grade1/grade1.cs.
See Grade Exercise.
As in a basic if-else statement, in the general format,
if ( condition1 ) {
statement-block-run-if-condition1-is-true;
}
else if ( condition2 ) {
statement-block-run-if-condition2-is-the-first-true;
}
else if ( condition3 ) {
statement-block-run-if-condition3-is-the-first-true;
}
// ...
else { // no condition!
statement-block-run-if-no condition-is-true;
}
exactly one of the statement blocks gets executed: If some condition is true, the first block following a true condition
is executed. If no condition is true, the else block is executed.
Here is a variation. Consider this fragment without a final else:
if (weight > 120) {
Console.WriteLine("Sorry, we can not take a suitcase that heavy.");
}
else if (weight > 50) {
Console.WriteLine("There is a $25 charge for luggage that heavy.");
}

This statement only prints one of two lines if there is a problem with the weight of the suitcase. Nothing is printed if
there is not a problem.
If the final else clause is omitted from the general if ... else if ... pattern above, at most one block after
a condition is executed: That is the block after the first true condition. If all the conditions are false, none of the
statement blocks will be executed.
It is also possible to embed if-else statements inside other if or if-else statements in more complicated patterns.

66 Chapter 5. Decisions
 Introduction to Computer Science in C#, Release 2.0

5.5.1 Sign Exercise

Write a program sign.cs to ask the user for a number. Print out which category the number is in: "positive",
"negative", or "zero".

5.5.2 Grade Exercise

Copy grade1/grade1.cs to grade2.cs in your own project. Modify grade2.cs so it has an equivalent version of
the letterGrade function that tests in the opposite order, first for F, then D, C, .... Hint: How many tests do you need to
do? 1
Be sure to run your new version and test with different inputs that test all the different paths through the program.
Be careful for edge cases: Test the grades on the “edge” of a change in the result.

5.5.3 Wages Exercise

Modify the wages1/wages1.cs or the wages2/wages2.cs example to create a program wages3.cs that assumes people
are paid double time for hours over 60. Hence they get paid for at most 20 hours overtime at 1.5 times the normal rate.
For example, a person working 65 hours with a regular wage of $10 per hour would work at $10 per hour for 40 hours,
at 1.5 * $10 for 20 hours of overtime, and 2 * $10 for 5 hours of double time, for a total of
10*40 + 1.5*10*20 + 2*10*5 = $800.
You may find wages2/wages2.cs easier to adapt than wages1/wages1.cs.

5.6 If-statement Pitfalls

5.6.1 Dangerous Semicolon

Regular statements must end with a semicolon. It turns out that the semicolon is all you need to have a legal statement:
;

We will see places that it is useful, but meanwhile it can cause errors: You may be hard pressed to remember to put
semicolons at the end of all your statements, and in response you may get compulsive about adding them at the end of
statement lines. Be careful NOT to put one at the end of a method heading or an if condition:
if ( x < 0); // WRONG PROBABLY!
Console.WriteLine(x);

This code is deadly, since it compiles and is almost surely not what you mean.
Remember indentation and newlines are only significant for humans. The two lines above are equivalent to:
if ( x < 0)
; // Do nothing as statement when the condition is true
Console.WriteLine(x); // past if statement - do it always

(Whenever you do need an empty statement, you are encouraged to put the semicolon all by itself on a line, as above.)
If you always put an open brace directly after the condition in an if statement, you will not make this error:
1 4 tests to distinguish the 5 cases, as in the previous version

5.6. If-statement Pitfalls 67

Introduction to Computer Science in C#, Release 2.0

if ( x < 0) {
Console.WriteLine(x);
}

Then even if you were to add a semicolon:

if ( x < 0) { ;
Console.WriteLine(x);
}

it would be a waste of a keystroke, but it would just be the first (empty) statement inside the block, and the writing
would still follow: The extra semicolon would have no effect.
The corresponding error at the end of a method heading will at least generate a compiler error, though it may appear
cryptic:
static void badSemicolon(int x);
{
x = x + 2;
// ...

This is another easy one to make and miss - just one innocent semicolon.

5.6.2 Match Wrong if With else

If you do not consistently put the substatements for the true and false choices inside braces, you can run into problems
from the fact that the else part of an if statement is optional. Even if you use braces consistently, you may well need
to read code that does not place braces around single statements. If C# understood indentation as in the recommended
formatting style (or as required in Python), the following would be OK:
if (x > 0)
if (y > 0)
Console.WriteLine("positive x and y");
else
Console.WriteLine("x not positive, untested y");

Unfortunately placing the else under the first if is not enough to make them go together (remember the C# compiler
ignores extra whitespace). The following is equivalent to the compiler, with the else apparently going with the second
if:
if (x > 0)
if (y > 0)
Console.WriteLine("positive x and y");
else
Console.WriteLine("x not positive, untested y");

The compiler is consistent with the latter visual pattern: an else goes with the most recent if that could still take
an else. Hence if x is 3 and y is -2, the else part is executed and statement printed is incorrect: the else clause is
only executed when x is positive and y (is tested and) is not positive. If you put braces everywhere to reinforce your
indentation, as we suggest, or if you only add the following one set of braces around the inner if statement:
if (x > 0) {
if (y > 0)
Console.WriteLine("positive x and y");
}
else
Console.WriteLine("x not positive, untested y");

68 Chapter 5. Decisions
 Introduction to Computer Science in C#, Release 2.0

then the braces enclosing the inner if statement make it impossible for the inner if to continue on to an optional
else part. The else must go with the first if. Now when the else part is reached, the statement printed will be
true: x is not positive, and the test of y is skipped.

5.6.3 Missing Braces

Another place you can fool yourself with nice indenting style is something like this. Suppose we start with a perfectly
reasonable
if (x > 0)
Console.WriteLine("x is: positive");

We may decide to avoid the braces, since there is just one statement that we want as the if-true part, but if we later
decide that we want this on two lines and change it to
if (x > 0)
Console.WriteLine("x is:");
Console.WriteLine(" positive");

We am not going to get the behavior we want. The word “positive” will always be printed.
If we had first taken a bit more effort originally to write
if (x > 0) {
Console.WriteLine("x is: positive");
}

then we could have split successfully into

if (x > 0) {
Console.WriteLine("x is:");
Console.WriteLine(" positive");
}

This way we do not have to keep worrying about this question when we revise: “Have I switched to multiple lines
after the if and need to introduce braces?”
The last two of the pitfalls mentioned in this section are fixed by consistent use of braces in the sub-statements of if
statements. Even with good use of braces, you still need to watch out for an incorrect semicolon after a condition.

5.7 Compound Boolean Expressions

To be eligible to graduate from Loyola University Chicago, you must have 120 credits and a GPA of at least 2.0. C#
does not use the word and. Instead it uses && (inherited from the C language). Then the requirement translates directly
into C# as a compound condition:
credits >= 120 && GPA >=2.0

This is true if both credits >= 120 is true and GPA >= 2.0 is true. A short example function using this would
be:
static void checkGraduation(int credits, double GPA)
{
if (credits >= 120 && GPA >=2.0) {
Console.WriteLine("You are eligible to graduate!");
}
else {

5.7. Compound Boolean Expressions 69

Introduction to Computer Science in C#, Release 2.0

Console.WriteLine("You are not eligible to graduate.");

}
}

The new C# syntax for the operator &&:

condition1 && condition2
The compound condition is true if both of the component conditions are true. It is false if at least one of the conditions
is false.
Suppose we want a C# condition that is true in the same situations as the mathematical expression: low < val < high.
Unfortunately the math is not a C# expression. The C# operator < is binary. In C# the statement above is equivalent to
(low < val) < high
comparing a Boolean result to high, and causing a compiler error. There is a C# version. Be sure to use this pattern:
low < val && val < high

Now suppose we want the opposite condition: that val is not strictly between low and high. There are several ap-
proaches. One is that val would be less than or equal to low or greater than or equal to high. C# translate or into
||, so a C# expression would be:
val <= low || val >= high
The new C# syntax for the operator ||:
condition1 || condition2
The compound condition is true if at least one of the component conditions are true. It is false if both conditions are
false.
Another logical way to express the opposite of the condition low < val < high is that it is not the case that low < val
&& val << high. C# translates not as !. Another way to state the condition would be
!(low < val && val < high)

The parentheses are needed because the ! operator has higher precedence than <.
A way to remember this strange not operator is to think of the use of ! in the not-equal operator: !=
The new C# syntax for the operator !:
! condition
This whole expression is true when condition is false, and false when condition is true.
Because of the precedence of !, you are often going to write:
!( condition )
Remember when such a condition is used in an if statement, outer parentheses are also needed:
if (!( condition )) {
We now have a lot of operators! Most of those in appendix Precedence of Operators have now been considered.
Compound Overkill: Look back to the code converting a score to a letter grade in Multiple Tests and if-else State-
ments. The condition before assigning the B grade could have been:
(score >= 80 && score < 90)

That would have totally nailed the condition, but it is overly verbose in the if .. else if ... code where it appeared:
Since you only get to consider a B as a grade if the grade was not already set to A, the second part of the compound
condition above is redundant.

70 Chapter 5. Decisions
 Introduction to Computer Science in C#, Release 2.0

5.7.1 Congress Exercise

A person is eligible to be a US Senator who is at least 30 years old and has been a US citizen for at least 9 years.
Write a version of a program congress.cs to obtain age and length of citizenship from the user and print out if a
person is eligible to be a Senator or not. A person is eligible to be a US Representative who is at least 25 years old
and has been a US citizen for at least 7 years. Elaborate your program congress.cs so it obtains age and length of
citizenship and prints whether a person is eligible to be a US Representative only, or is eligible for both offices, or is
eligible for neither.
This exercise could be done by making an exhaustive treatment of all possible combinations of age and citizenship.
Try to avoid that. (Note the paragraph just before this exercise.)

5.7. Compound Boolean Expressions 71

Introduction to Computer Science in C#, Release 2.0

72 Chapter 5. Decisions
 CHAPTER

SIX

WHILE LOOPS

6.1 While-Statements

We have seen that the sequential flow of a program can be altered with function calls and decisions. The last important
pattern is repetition or loops. There are several varieties. The simplest place to start is with while loops.
A C# while loop behaves quite similarly to common English usage. If you hear
While your tea is too hot, add a chip of ice.
Presumably you would test your tea. If it were too hot, you would add a little ice. If you test again and it is still too
hot, you would add ice again. As long as you tested and found it was true that your tea was too hot, you would go back
and add more ice. C# has a similar syntax:
while ( condition )
statement
As with an if statement we will generally assume a compound statement, after the condition, so the syntax will
actually be:
while ( condition ) {
statement(s)
}
Setting up the English example in a similar format would be:
while ( your tea is too hot ) {
add a chip of ice
}
To make things concrete and numerical, suppose the following: The tea starts at 115 degrees Fahrenheit. You want
it at 112 degrees. A chip of ice turns out to lower the temperature one degree each time. You test the temperature
each time, and also print out the temperature before reducing the temperature. In C# you could write and run the code
below, saved in example program cool.cs:
We added a final line after the while loop to remind you that execution follows sequentially after a loop completes.
It is extremely important to totally understand how the flow of execution works with loops. One way to follow it
closely is to make a table with a line for each instruction executed, keeping track of all the variables. We call this
playing computer.
Each row shows the line number of the start of the next instruction executed, and the values of all the variables after
the instruction is executed. The important thing to see with loops is that the same line can be executed over and over,
but with different variable values. We leave a column for the line number, each variable that is involved (particularly
any that change) and a place for comments about what is happening. The comment line can be used any time it is

73
Introduction to Computer Science in C#, Release 2.0

helpful. If should be used in particular when something is printed and when something is returned, since neither of
these important actions appear in the variable list.
If you play computer and follow the path of execution, you could generate the following table. Remember, that each
time you reach the end of the block after the while heading, execution returns to the while heading for another test:
Line temperature Comment
1 115
2 115 > 112 is true, do loop
3 prints 115
4 114 115 - 1 is 114, loop back
2 114 > 112 is true, do loop
3 prints 114
4 113 114 - 1 is 113, loop back
2 113 > 112 is true, do loop
3 prints 113
4 112 113 - 1 is 112, loop back
2 112 > 112 is false, skip loop
6 prints that the tea is cool
Each time the end of the loop body block is reached, execution returns to the while loop heading for another test.
When the test is finally false, execution jumps past the indented body of the while loop to the next sequential
statement.
The biggest trick with a loop is to make the same code do the next thing you want each time through. That generally
involves the use of variables that are modified for each successive time through the loop.
initialization
while ( continuationCondition ) {
do main action to be repeated
prepare variables for the next time through the loop
}
The simple first example follows this pattern directly. Note that the variables needed for the test of the condition must
be set up both in the initialization and inside the loop (often at the very end). Without a change inside the loop, the
loop would run forever!
It is a big deal for beginning students, how to manage all this in general. We will see a number of common patterns in
lots of practice. We will use the term successive modification loop for loops following this pattern.
Test yourself: Follow the code. Figure out what is printed. If it helps, get detailed and play computer:
Check yourself by running the example program test_while1/test_while1.cs.

Note: In C#, while is not used quite like in English. In English you could mean to stop as soon as the condition
you want to test becomes false. In C# the test is only made when execution for the loop starts (or starts again), not in
the middle of the loop.

Predict what will happen with this slight variation on the previous example, switching the order in the loop body.
Follow it carefully, one step at a time.
Check yourself by running the example program test_while2/test_while2.cs.
The sequence order is important. The variable i is increased before it is printed, so the first number printed is 6.
Another common error is to assume that 10 will not be printed, since 10 is past 9, but the test that may stop the loop is
not made in the middle of the loop. Once the body of the loop is started, it continues to the end, even when i becomes
10.

74 Chapter 6. While Loops

 Introduction to Computer Science in C#, Release 2.0

Line i Comment
1 4
2 4 < 9 is true, do loop
3 6 4+2=6
4 print 6
2 6 < 9 is true, do loop
3 8 6+2= 8
4 print 8
2 8 < 9 is true, do loop
3 10 8+2=10 No test here
4 print 10
2 10 < 9 is false, skip loop
You should be able to generate a table like the one above, following the execution of one statement at a time. You are
playing through the role of the computer in detail. We will refer to this later as playing computer. As code gets more
complicated, particularly with loops, this is an important skill.
Problem: Write a program with a while loop to print:
10
9
8
7
6
5
4
3
2
1
Blastoff!
Analysis: We have seen that we can produce a regular sequence of numbers in a loop. The “Blastoff!” part does not
fit the pattern, so it must be a separate part after the loop. We need a name for the number that decreases. It can be
time. Remember the general rubric for a while loop:
initialization
while ( continuationCondition ) {
do main action to be repeated
prepare variables for the next time through the loop
}
You can consider each part separately. Where to start is partly a matter of taste.
The main thing to do is print the time over and over. The initial value of the time is 10. We are going to want to keep
printing until the time is down to 1, so we continue while the time is at least 1, meaning the continuationCondition can
be time >= 1, or we could use time > 0.
Finally we need to get ready to print a different time in the next pass through the loop. Since each successive time is
one less than the previous one, the preparation for the next value of time is: time = time - 1.
Putting that all together, and remembering the one thing we noted to do after the loop, we get blastoff/blastoff.cs:
Look back and see how we fit the general rubric. There are a bunch of things to think about with a while loop, so it
helps to go one step at a time, thinking of the rubric and the specific needs of the current problem.
There are many different (and more exciting) patterns of change coming for loops, but the simple examples so far get
us started.

6.1. While-Statements 75
Introduction to Computer Science in C#, Release 2.0

Looking ahead to more complicated and interesting problems, here is a more complete list of questions to ask yourself
when designing a function with a while loop:
• What variables do We need?
• What needs to be initialized and how? This certainly includes any variable tested in the condition.
• What is the condition that will allow the loop to continue?
• What is the code that should only be executed once? What action do I want to repeat? Where does the repetition
come in the overall sequence of operations?
• How do I write the action so I can modify it for the next time through the loop?
• What code is needed to do modifications to make the same code work the next time through the loop?
• Have I thought of variables needed in the middle and declared them; do other things need initialization?
• Will the continuation condition eventually fail? Be sure to think about this!
• Separate actions to be done once before the repetition (code before the loop) from repetitive actions (in the loop)
from actions not repeated, but done after the loop (code after the loop). Missing this distinction is a common
error!

6.1.1 Sum To n

Let us write a function to sum the numbers from 1 to n:

/// Return the sum of the numbers from 1 through n.
static int SumToN(int n)
{
...
}

For instance SumToN(5) calculates 1 + 2 + 3 + 4 + 5 and returns 15. We know how to generate a sequence of integers,
but this is a place that a programmer gets tripped up by the speed of the human mind. You are likely so quick at this
that you just see it all at once, with the answer.
In fact, you and the computer need to do this in steps. To help see, let us take a concrete example like the one above
for SumToN(5), and write out a detailed sequence of steps like:
3 = 1 + 2
6 = 3 + 3
10 = 6 + 4
15 = 10 + 5

You could put this in code directly for a specific sum, but if n is general, we need a loop, and hence we must see a
pattern in code that we can repeat.
Each of the second terms in the additions is a successive integer, that we can generate. Starting in the second line,
the first number in each addition is the sum from the previous line. Of course the next integer and the next partial
sum change from step to step, so in order to use the same code over and over we will need changeable variables, with
names. We can make the partial sum be sum and we can call the next integer i. Each addition can be in the form:
sum + i

We need to remember that result, the new sum. you might first think to introduce such a name:
newSum = sum + i;

76 Chapter 6. While Loops

 Introduction to Computer Science in C#, Release 2.0

This will work. We can go through the while loop rubric:

The variables are sum, newSum and i.
To evaluate
newSum = sum + i;
the first time in the loop, we need initial values for sum and i. Our concrete example leads the way:
int sum = 1, i = 2;

We need a while loop heading with a continuation condition. How long do we want to add the next i? That is for
all the value up to and including n:
while (i <= n) {

There is one more important piece - making sure the same code
newSum = sum + i;
works for the next time through the loop. We have dealt before with the idea of the next number in sequence:
i = i + 1;

What about sum? What was the newSum on one line becomes the old or just plain sum on the next line, so we can
make an assignment:
sum = newSum:

All together we calculate the sum with:

int sum = 1, i = 2;
while (i <= n) {
int newSum = sum + i;
i = i + 1;
sum = newSum:
}

This exactly follows our general rubric, with preparation for the next time through the loop at the end of the loop. We
can condense it in this case: Since newSum is only used once, we can do away with it, and directly change the value
of sum:
int sum = 1, i = 2;
while (i <= n) {
sum = sum + i;
i = i + 1;
}

Finally this was supposed to fit in a function. The ultimate purpose was to return the sum, which is the final value of
the variable sum, so the whole function is:
/// Return the sum of the numbers from 1 through n.
static int SumToN(int n) // line 1
{
int sum = 1, i = 2; // 2
while (i <= n) { // 3
sum = sum + i; // 4
i = i + 1; // 5
}
return sum; // 6
}

6.1. While-Statements 77
Introduction to Computer Science in C#, Release 2.0

The comment before the function definition does not give a clear idea of the range of possible values for n. How
small makes sense for the comment? What actually works in the function? The smallest expression starting with 1
would just be 1: (n = 1). Does that work in the function? You were probably not thinking of that when developing the
function! Now look back now at this edge case. You can play computer on the code or directly test it. In this case the
initialization of sum is 1, and the body of the loop never runs (2 <= 1 is false). The function execution jumps right to
the return statement, and does return 1, and everything is fine.
Also you should check the program in a more general situation, say with n being 4. You should be able to play
computer and generate this table, using the line numbers shown in comments at the end of lines, and following one
statement of execution at a time. We only make entries where variables change value.
Line i sum Comment
1 assume 4 is passed for n
2 2 1
3 2<=4: true, enter loop
4 3 1+2=3
5 3 2+1=3, bottom of loop
3 3<=4: true
4 6 3+3=6
5 4 3+1=4, bottom of loop
3 4<=4: true
4 10 6+4=10
5 5 4+1=5, bottom of loop
3 5<=4: false, skip loop
6 return 10
Now about large n....
With loops we can make programs run for a long time. The time taken becomes an issue. In this case we go though
the loop n-1 times, so the total time is approximately proportional to n. We write that the time is O(n), spoken “oh of
n”, or “big oh of n” or “order of n”.
Computers are pretty fast, so you can try the testing program sum_to_n_test/sum_to_n_test.cs and it will go by so
fast, that you will hardly notice. Try these specific numbers in tests: 5, 6, 1000, 10000, 98765. All look OK? Now try
66000. On many systems you will get quite a surprise! This is the first place we have to deal with the limited size of
the int type. On many systems the limit is a bit over 2 billion. You can check out the size of int.MaxValue in
csharp. The answer for 66000, and also 98765, is bigger than the upper limit. Luckily the obviously wrong negative
answer for 66000 pops out at you. Did you guess before you saw the answer for 66000, that there was an issue for
98765? It is a good thing that no safety component in a big bridge was being calculated! It is a big deal that the system
fails silently in such situations. Think how large the data may be that you deal with!
Now look at and run sum_to_n_long/sum_to_n_long.cs. The sum is a long integer here. Check out in csharp how
big a long can be (long.MaxValue). This version of the program works for 100000 and for 98765. We can get
correct answers for things that will take perceptible time. Try working up to 1 billion (1000000000, nine 0’s). It takes
a while: O(n) can be slow!
By hand it is a lot slower, unless you totally change the algorithm: There is a classic story about how a calculation like
this was done in grade school (n=100) by the famous mathematician Gauss. His teacher was trying to keep him busy.
Gauss discovered the general, exact, mathematical formula:
1 + 2 + 3 + ... + n = n(n+1)/2.
That is the number of terms (n), times the average term (n+1)/2.
Our loop was instructive, but not the fastest approach. The simple exact formula takes about the same time for any n.
(That is as long as the result fits in a standard type of computer integer!) This is basically constant time. In discussing
how the speed relates to the size of n, we say it is O(1). The point is here that 1 is a constant. The time is of constant
order.

78 Chapter 6. While Loops

 Introduction to Computer Science in C#, Release 2.0

We can write a ridiculously short function following Gauss’s model. Here we introduce the variable average, as in the
motivation for Gauss’s answer:
Run the example program containing it: sum_to_n_long_bad/sum_to_n_long_bad.cs.
Test it with 5, and then try 6. ???
“Ridiculously short” does not imply correct! The problem goes back to the fact that Gauss was in math class and you
are doing Computer Science. Think of a subtle difference that might come in here: Though (n+1)/2 is fine as math,
recall the division operator does not always give correct answers in C#. You get an integer answer from the integer
(or long) operands. Of course the exact mathematical final answer is an integer when adding integers, but splitting it
according to Gauss’s motivation can put a mathematical non-integer in the middle.
The C# fix: The final answer is clearly an integer, so if we do the division last, when we know the answer will be an
integer, things should be better:
return n*(n+1)/2;

Here is a shot at the whole function:

Run the example program containing it: sum_to_n_long_bad2/sum_to_n_long_bad2.cs.
Test it with 5, and then try 6. Ok so far, but go on to long integer range: try 66000 that messed us up before. ??? You
get an answer that is not a multiple of 1000: not what we got before! What other issues do we have between math and
C#?
Further analysis: To make sure the function always worked, it made sense to leave the parameter n an int. The
function would not work with n as the largest long. The result can still be big enough to only fit in a long, so
the return value is a long. All this is reasonable but the C# result is still wrong! Look deeper. While the result
of n*(n+1)/2 is assigned to a long variable, the calculation n*(n+1)/2 is done with ints not mathematical
integers. By the same general type rule that led to the (n+1)/2 error earlier, these operations on ints produce an int
result, even when wrong.
We need to force the calculation to produce a long. In the correct looping version sum was a long, and that forced
all the later arithmetic to be with longs. Here are two variations that work:
long nLong = n;
return nLong*(nLong+1)/2;

or we can avoid a new variable name by Casting to long, converting the first (left) operand to long, so all the later
left-to-right operations are forced to be long:
return (long)n*(n+1)/2;

You can try example sum_to_n_long_quick/sum_to_n_long_quick.cs to finally get a result that is dependably fast and
correct.
Important lessons from this humble summation problem:
• Working and being efficient are two different things in general.
• Math operations and C# operations are not always the same. Knowing this in theory is not the same as remem-
bering it in practice!
Further special syntax that only makes sense in any kind of loop is discussed in Break and Continue, after we introduce
the last kind of loop.

6.1. While-Statements 79
Introduction to Computer Science in C#, Release 2.0

6.2 While-Statements with Sequences

We will process many sequences or collections. At this point the only collection we have discussed is a string - a
sequence of characters that we can index. Consider the following silly function description and heading as a start:
OneCharPerLine("bug") would print:
b
u
g
We are accessing one character at a time. We can do that with the indexing notation. Thinking concretely about
the example above, we are looking to print, s[0], s[1], s[2]. This requires a loop. For now our only option is a
while loop. We can follow our basic rubric, one step at a time: The index is changing in a simple repetitive sequence.
We can call the index i. Its initial value is clearly 0. That is our initialization. We need a while loop continuation
condition. For the 3-character string example, the last index above is 2. In general we want all the characters. Recall
the index of the last character is the length - 1, or with our parameter s, s.Length - 1 The while loop condition
needs to allow indices through s.Length - 1. We could write a condition with <= or more concisely:
while (i < s.Length) {

In the body of the loop, the main thing is to print the next character, and the next character is s[i]:
Console.WriteLine(s[i]);

We also need to remember the part to get ready for the next time through the loop. We have dealt with regular sequence
of values before. We change i with:
i = i+1;

This change is so common, there is a simpler syntax:

i++;

This increases the value of the numeric variable i by 1. (The reverse is i--;) 1
So all together:
You can test this with example char_loop1/char_loop1.cs.
This is a very common pattern. We could do anything we want with each individual character, not just print it.

PrintVowels

Let us get more complicated. Consider the function described:

For instance PrintVowels(“hello”) would print:
e
o
We have seen that we can go through the whole string and do the same thing each time through the loop, using s[i]
in some specific way.
This new description seems like a problem. We do not appear to want to do the same thing each time: We only want to
print some of the characters. Again your eyes and mind are so fast, you likely miss what you need to do when you go
1 To be complete, the statements c = c + 1; and c++; are not always equivalent.

In c++ the type of c must be integral, but not necessarily int. It could be a smaller type, like char.
The c++ could not be replaced by c = c + 1, but you could use c = (char)(c + 1): The int literal 1 forces the sum expression to be
an int, which must be cast back to a char to be assigned to c. Similarly with the -- operator.

80 Chapter 6. While Loops

 Introduction to Computer Science in C#, Release 2.0

through PrintVowels by hand. Your eyes let you just grab the vowels easily, but think, what is actually happening?
You are checking each character to see if it is a vowel, and printing it if it is: You are doing the same thing each time -
testing if the character is a vowel. The pseudocode is
if (s[i] is a vowel) {
print s[i]
}

We do want to do this each time through the loop. We can use a while statement.
Next problem: convert the pseudocode “s[i] is a vowel” to C#.
There are multiple approaches. The one you get by following your nose is just to consider all the cases where it is true:
s[i] == ’a’
s[i] == ’e’
s[i] == ’i’
s[i] == ’o’
s[i] == ’u’

How do you combine them into a condition? The letter can be a or e or i or o or u. We get the code:
That has a long condition! Here is a nice trick to shorten that: We want to check if a character is in a group of letters.
We have already seen the string method IndexOf. Recall we can use it to see if a character is in or not in a string. We
can use "aeiou".IndexOf(s[i]). We do not care where s[i] comes in the string of vowels. All we care is
that "aeiou".IndexOf(s[i]) >= 0.
This is still a bit of a mouthful. Often it is just important if a character or string is contained in another string, not where
it appears, so it is easier to use the string method Contains. Though IndexOf takes either a string or a character as
parameter, Contains only takes a string. There is a nice quick idiom to convert anything to a string: use ""+. The
condition could be "aeiou".Contains(""+s[i]). This adds the string version of s[i] to the empty string.
The function is still not as general as it might be: Only lowercase vowels are listed. We could do something with
ToLower, or just use the condition: "aeiouAEIOU".Contains(""+s[i])
This variation is in example vowels2/vowels2.cs.

IsDigits

Consider a variation, determining if all the characters in a string are vowels. We could work on that, but it is not very
useful. Instead let us consider if all the characters are digits. This is a true-false question, so the function to determine
this would return a Boolean result:
There are several ways to check if a character is a digit. We could use the Contains idiom from above, but here
is another option: The codes for digits are sequential, and since characters are technically a kind of integer, we can
compare: The character s[i] is a digit if it is in the range from ’0’ to ’9’, so the condition can be written:
’0’ <= s[i] && s[i] <= ’9’

Similarly the condition s[i] is not a digit, can be written negating the compound condition as in Compound Boolean
Expressions:
s[i] < ’0’ || s[i] > ’9’

If you think of going through by hand and checking, you would check through the characters sequentially and if you
find a non-digit, you would want to remember that the string is not only digits.
One way to do this is have a variable holding an answer so far:

6.2. While-Statements with Sequences 81

Introduction to Computer Science in C#, Release 2.0

bool allDigitsSoFar = true;

Of course initially, you have not found any non-digits, so it starts off true. As you go through the string, you want to
make sure that answer is changed to false if a non-digit is encountered:
if (’0’ > s[i] || s[i] > ’9’) {
allDigitsSoFar = false;
}

When we get all the way through the string, the answer so far is the final answer to be returned:
/// Return true if s contains one or more digits
/// and nothing else. Otherwise return false.
static bool IsDigits(string s)
{
bool allDigitsSoFar = true;
int i = 0;
while (i < s.Length) {
if (s[i] < ’0’ || s[i] > ’9’) {
allDigitsSoFar = false;
}
i++;
}
return allDigitsSoFar;
}

Remember something to always consider: edge cases. In the description it says it is true for a string of one or more
digits.
Check examples of length 1 and 0. Length 1 is fine, but it fails for the empty string, since the loop is skipped and the
initial answer, true is returned.
There are many ways to fix this. We will know right up front that the answer is false if the length is 0, and we could
immediately set allDigitsSoFar to false. We would need to change the initialization so it checks the length and
chooses the right value for allDigitsSoFar, true or false. Since we are selecting between two values, an if
statement should occur to you:
bool allDigitsSoFar;
if (s.Length > 0) {
allDigitsSoFar = true;
}
else {
allDigitsSoFar = false;
}

If we substitute this initialization for allDigitsSoFar, the code will satisfy the edge case, and the code will always
work. Still, this code can be improved:
Examine the if statement more closely:
if the condition is true, allDigitsSoFar is true;
if the condition is false, allDigitsSoFar is false;
See the symmetry: the value assigned to allDigitsSoFar is always the value of the condition.
A much more concise and still equivalent initialization is just:
bool allDigitsSoFar = (s.Length > 0);

In more generality this conciseness comes from the fact that it is a Boolean value that you are trying to set, based on a
Boolean condition: You do not need to do that with an if statement! You just need an assignment statement! If you

82 Chapter 6. While Loops

 Introduction to Computer Science in C#, Release 2.0

use an if statement in such a situation, you being verbose and marking yourself as a novice.
It could even be slightly more concise: The precedence of assignment is very low, lower than the comparison >, so the
parentheses could be omitted. We think the code is easier to read with the parentheses left in, as written above, and
below.
The whole function would be:
You can try this code in example check_digits1/check_digits1.cs.
We are not done. This code is still inefficient. If an early character in a long string is not a digit, we already know the
final answer, but this code goes through and still checks all the other characters in the string! People checking by hand
would stop as soon as they found a non-digit. We can do that in several ways with C#, too. Since this is a function, and
we would know the final answer where we find a non-digit, the simplest thing is to use the fact that a return statement
immediately terminates the function (even if in a loop).
Instead of setting a variable to false to later be returned, we can return right away, using the loop:
while (i < s.Length) {
if (s[i] < ’0’ || s[i] > ’9’) {
return false;
}
i++;
}

What if the loop terminates normally (no return from inside)? That means no non-digit was found, so if there are any
characters at all, they are all digits. There are one or more digits as long as the string length is positive. Again we do
not need an if statement for a check. Look in the full code for the function:
The full code with a Main testing program is in example check_digits2/check_digits2.cs.
Returning out of a loop is a good pattern to remember when you are searching for something, and you know the final
answer for your function as soon as you find it.

6.2.1 Exercise

Duplicate Character Exercise

Create a file double_char_test.cs, and write and test a function with the documentation and heading below:
/// If two consecutive characters in s are the same, return true.
/// Return false otherwise. Examples:
/// HasDoubleChar("bigfoot") and HasDoubleChar("aaah!") are true;
/// HasDoubleChar("treated") and HasDoubleChar("haha!") are false.
static bool HasDoubleChar(string s)

6.3 Interactive while Loops

Next we consider a particular form of while loops: Interactive while loops involve input from the user each time
through. We consider them now for three reasons:
• Interactive while loops have one special ‘gotcha’ worth illustrating.
• We will illustrate some general techniques for understanding and developing while loops.
• As a practical matter, we can greatly improve the utility input functions we have been using, and add some more.

6.3. Interactive while Loops 83

Introduction to Computer Science in C#, Release 2.0

We already have discussed the PromptInt function. The user can choose any int. Sometimes we only want an integer
in a certain range. One approach is to not accept a bad value, but make the user repeat trying until explicitly entering
a value in the right range. In theory the user could make errors for some time, so a loop makes sense. For instance we
might have a slow user, and there could be an exchange like the following when you want a number from 0 to 100.
For illustration, user input is shown in boldface:
Enter a score: (0 through 100) 233
233 is out of range!
Enter a score: (0 through 100) 101
101 is out of range!
Enter a score: (0 through 100) -1
-1 is out of range!
Enter a score: (0 through 100) 100
and the value 100 would be accepted.
This is a well-defined idea. A function makes sense. Its heading includes a prompt and low and high limits of the
allowed range:
For example to generate sequence above, the call would be:
PromptIntInRange("Enter a score: (0 through 100) ", 0, 100)
There is an issue with the common term “loop” in programming. In normal English, a loop has no beginning and no
end, like a circle. C# loops have a sequence of statements with a definite beginning and end.
Consider the sequence above in pseudocode.
Input a number with prompt (233)
Print error message
Input a number with prompt (101)
Print error message
Input a number (-1)
Print error message
Input a number with prompt (100)
Return 100
We can break this into a repeating pattern in two ways. The most obvious is the following, with three repetitions of a
basic pattern, with the last two line not in the same pattern (so they would go after the loop). :
Input a number with prompt (233)
Print error message

Input a number with prompt (101)

Print error message

Input a number (-1)

Print error message

Input a number with prompt (100)

Return 100
Another choice, since you can split a loop at any point, would be the following, with the first and last lines not in the
pattern that repeats three times in the middle:
Input a number with prompt (233)

84 Chapter 6. While Loops

 Introduction to Computer Science in C#, Release 2.0

Print error message

Input a number with prompt (101)

Print error message

Input a number (-1)

Print error message

Input a number with prompt (100)

Return 100
When you consider while loops, there is a problem with the first version: Before the first pass through the loop and
at the end of the block of code in the body of the loop, you must be able to run the test in the while heading. We will
be testing the latest input from the user.
It is the second version that has us getting new input before the first loop and at the end of each loop!
Now we can think more of the basic process to turn this into a C# solution: What variables do we need? We will call
the user’s response number.
What is the test in the while loop heading? The easiest thing to think of is that we are done when we get something
correct. That, however, is a termination condition. We need to reverse it to get the continuation condition, that the
answer is out of range. There are two ways to be out of range:
number < lowLim
number > highLim

How do we combine them? Either one rules out a correct answer, so number is out of range if too high OR too low.
Remember the C# symbolism for “or”: ||:
while (number < lowLim || number > highLim) {

Following the sequence in the concrete example we had above, we can see how to put things together. We need to get
input from the user before first beginning the while loop, so we immediately have something to test in the while
heading’s condition.
Do not reinvent the wheel! We can use our earlier general PromptInt function. It needs a prompt. As a first version,
we can use the parameter prompt:
int number = PromptInt(prompt);

That is the initialization step before the loop.

If we get into the body of the loop, it means there is an error, and the concrete example indicates we print a warning
message. The concrete example also shows another step in the loop, asking the user for input. It is easy to think
“I already have the code included to read a value from the user, so there is nothing really to do.”
WRONG! The initialization code with the input from the user is before the loop. C# execution approaches the test in
the while headings from two places: the initialization and coming back from the bottom of the loop. To get a new
value to test, we must repeat getting input from the user at the bottom of the loop body.
You might decide to be quick and just copy the initialization line into the bottom of the loop (and indent it):
int number = PromptInt(prompt);

Luckily you will get a compiler error in that situation, avoiding more major troubleshooting: The complete copy of
the line copies the declaration part as well as the assignment part, and the compiler sees the declaration of number

6.3. Interactive while Loops 85

Introduction to Computer Science in C#, Release 2.0

already there from the scope outside the while block, and complains.
Hence copy the line, without the int declaration:
number = PromptInt(prompt);

When the loop condition becomes false, and you get past the loop, you have a correct value in number. You have
done all the hard work. Do not forget to return it at the end.
You can try this full example, input_in_range1/input_in_range1.cs. Look at it and then try compiling and running.
Look at the Main code. It is redundant - the limits are written both in the prompt and in the parameters. We can do
better. In general we endeavor to supply data only once, and let the program use it in several places if it needs to. In
this case we can complete the last part of the prompt automatically in the code:
There are two approaches here: The caller could give a more explicit prompt. Since the limits are given as parameters,
anyway, we prefer to have the program elaborate the prompt. If the limits are -10 and 10, automatically add to the
prompt something like (-10 through 10).
We could use
Console.Write(" ({0} through {1}) ", lowLim, highLim);

but we need the code twice, and it is quite a mouthful....

Thus far we have only seen the use of format strings when immediately printing with Console.Write (or
WriteLine). Here we would like to generate a string, for use later.
We introduce the C# library function string.Format, which does just what we want: The parameters have the
same form as for Console.Write, but the formatted string is returned.
Here is a revised version, in example input_in_range2/input_in_range2.cs, without redundancy in the prompts in
Main:
The only caveat with string.Format is that there is no special function corresponding to
Console.WriteLine, with an automatic terminating newline. You can generate a newline with string.Format:
Remember the escape code "\n". Put it at the end to go on to a newline.
This time around we did the user input correctly, with the request for new input repeated at the end of the loop. That
repetition is easy to forget. Before we see what happens when you forget, note:

Warning: A while loop may be written so the continuation condition is always true, and the loop never stops
by itself. This is an infinite loop. In practice, in many operating environments, particularly where you are getting
input from the user, you can abort the execution of a program in an infinite loop by entering Ctrl-C.

In particular you get an infinite loop if you fail to get new input from the user at the end of the loop. The
condition uses the bad original choice forever. Here is the loop in the mistaken version, from example in-
put_in_range2_bad/input_in_range2_bad.cs:
You can run the program. Remember Ctrl-C ! There are two tests in Main. If you give a legal answer immediately
in the first test, it works fine (never getting into the loop body). If you give a bad input in the second test, you see that
you can never fix it! Remember Ctrl-C !
A more extreme abort is to close the entire console/terminal window running the program.

6.3.1 Agree Function Exercise

Save example test_agree_stub/test_agree.cs in a project of your own.

Yes-no (true/false) questions are common. How might you write an input utility function Agree? You can speed
things up by considering only the first letter of responses. Assume that it is important that the user enter correctly, you

86 Chapter 6. While Loops

 Introduction to Computer Science in C#, Release 2.0

should consider three categories of answer: ones accepted as true, ones accepted as false, and ambiguous ones. You
need to allow for the possibility that the user keeps giving ambiguous answers....

6.3.2 Interactive Sum Exercise

Write a program sum_all.cs that prompts the user to enter numbers, one per line, ending with a line containing 0,
and keep a running sum of the numbers. Only print out the sum after all the numbers are entered (at least in your final
version).

6.3.3 Safe Whole Number Input Exercise

Save example test_input_whole_stub/test_input_whole.cs as a project of your own. The code should test a function
PromptWhole, as described below.
There is an issue with reading in numbers with the PromptInt function. If you make a typo and enter something that
cannot be converted from a string to the right kind of number, a naive program will bomb. This is avoidable if you
test the string and repeat the input if the string is illegal. Places where more complicated tests for illegality are needed
are considered in Safe PromptInt and PromptDouble Exercise. For now we just consider reading in whole numbers
(integers greater than or equal to 0). Note that such a number is written as just a sequence of digits. Follow the
interactive model of PromptIntInRange, looping until the user enters something that is legal: in this case, all digits.
The stub code already includes the earlier function IsDigits.

6.4 Short-Circuiting && and ||

Follow along with the following silly, but illustrative csharp sequence:
csharp> int x = 5, y = 2, z = 1;
csharp> y/x > z && x != 0;
false
csharp> x = 2; y = 5;
csharp> y/x > z && x != 0;
true

The compound condition includes x != 0, so what happens if we change x to 0 and try the condition again. Will
you get false?
csharp> x = 0;
csharp> y/x > z && x != 0;
System.DivideByZeroException: Division by zero
...

No, one of the parts involves dividing by zero, and you see the result. What if we swap the two conditions to get the
logically equivalent
csharp> x != 0 && y/x > z;
false

Something is going on here besides pure mathematical logic. Remember the final version in IsDigits. We did not
need to continue processing when we knew the final answer already. The && and || operators work the same way,
evaluating from left to right. If x != 0 is false, then x != 0 && y/x > z starts off being evaluated like
false && ??. We do no need the second part evaluated to know the overall result is false, so C# does not
evaluate further. This behavior has acquired the jargon short-circuiting. Many computer languages share this feature.

6.4. Short-Circuiting && and || 87

Introduction to Computer Science in C#, Release 2.0

It also applies to ||. In what situation do you know what the final result is after evaluating the first condition? In this
case you know:
true || ??

evaluates to true. Continuing with the same csharp sequence above (where x is 0, y is 5, and z is 1):
csharp> x == 0 || y/x > z;
true

The division by 0 in the second condition never happens. It is short-circuited.

For completeness, try the other order:
csharp> y/x > z || x == 0;
System.DivideByZeroException: Division by zero
...

This idea is useful in the Agree function, where you want to deal with the first character in the user’s answer.
In situations where you want to test conditionThatWillBombWithBadData, you want to avoid causing an Exception.
When there is good data, you want the result to actually come from conditionThatWillBombWithBadData. There are
two cases, however, depending on what result you want if the data for this condition is bad, so you cannot evaluate it:
• If you want the result to be false with bad data for the dangerous condition, use
falseConditionIfDataBad && conditionThatWillBombWithBadData
• If you want the result to be true with bad data for the dangerous condition, use
trueConditionIfDataBad || conditionThatWillBombWithBadData

6.5 While Examples

Todo
“bisection method”

6.5.1 Savings Exercise

The idea here is to see how many years it will take a bank account to grow to at least a given value, assuming a fixed
annual interest. Write a program savings.cs. Prompts the user for three numbers: an initial balance, the annual
percentage for interest as a decimal. like .04 for 4%, and the final balance desired. Print the initial balance, and the
balance each year until the desired amount is reached. Round displayed amounts to two decimal places, as usual.
The math: The amount next year is the amount now times (1 + interest fraction), so if I have $500 now and the interest
rate is .04, I have $500*(1.04) = $520 after one year, and after two years I have, $520*(1.04) = $540.80. If I enter into
the program a $500 starting balance, .04 interest rate and a target of $550, the program prints:
500.00
520.00
540.80
563.42

88 Chapter 6. While Loops

 Introduction to Computer Science in C#, Release 2.0

6.5.2 Strange Sequence Exercise

Save the example program strange_seq_stub/strange_seq.cs in a project of your own.

There are three functions to complete. Do one at a time and test.
Jump: First complete the definitions of function Jump. For any integer n, Jump(n) is n/2 if n is even, and 3*n+1
if n is odd. In the Jump function definition use an if-else statement. Hint 2
PrintStrangeSequence: You can start with one number, say n = 3, and keep applying the Jump function to the
last number given, and see how the numbers jump around!
Jump(3) = 3*3+1 = 10; Jump(10) = 10/2 = 5;
Jump(5) = 3*5+1 = 16; Jump(16) = 16/2 = 8;
Jump(8) = 8/2 = 4; Jump(4) = 4/2 = 2;
Jump(2) = 2/2 = 1

This process of repeatedly applying the same function to the most recent result is called function iteration. In this case
you see that iterating the Jump function, starting from n=3, eventually reaches the value 1.
It is an open research question whether iterating the Jump function from an integer n will eventually reach 1, for
every starting integer n greater than 1. Researchers have only found examples of n where it is true. Still, no general
argument has been made to apply to the infinite number of possible starting integers.
In the PrintStrangeSequence you iterate the Jump function starting from parameter value n, until the result is 1.
CountStrangeSequence: Iterate the Jump function as in PrintStrangeSequence. Instead of printing each
number in the sequence, just count them, and return the count.

6.6 More String Methods

Before we do more elaborate things with strings, some more string methods will be helpful. Be sure you are familiar
with the earlier discussion of strings in Basic String Operations.
Play with the new string methods in csharp!
This variation of IndexOf has a second parameter:
int IndexOf(string target, int start) Returns the index of the beginning of the first occurrence of
the string target in this string object, starting at index start or after. Returns -1 if target is not found.
Example:
csharp> string state = "Mississippi";
csharp> print("01234567890\n"+state) // to see indices
01234567890
Mississippi
csharp> state.IndexOf("is", 0); // same as state.IndexOf("is");
1
csharp> state.IndexOf("is", 2);
4
csharp> state.IndexOf("is", 5);
-1
csharp> state.IndexOf("i", 5);
7

string Trim() Returns a string formed from this string object, but with leading and trailing whitespace removed.
Example:
2 If you divide an even number by 2, what is the remainder? Use this idea in your if condition.

6.6. More String Methods 89

Introduction to Computer Science in C#, Release 2.0

csharp> string s = "\n 123 ";

csharp> "#" + s + "#";
#
123 #
csharp> "#" + s.Trim() + "#";
#123#

string Replace(string target, string replacement) Returns a string formed from this string by
replacing all occurrences of the substring target by replacement. Example:
csharp> string s = "This is it!";
csharp> s.Replace(" ", "/");
"This/is/it!"
csharp> s.Replace("is", "at");
"That at it!"
csharp> "oooooh".Replace("oo", "ah");
"ahahoh"

bool StartsWith(string prefix) Returns true if this string object starts with string prefix, and
false otherwise. Example:
csharp> "-123".StartsWith("-");
true
csharp> "downstairs".StartsWith("down");
true
csharp> "1 - 2 - 3".StartsWith("-");
false

bool EndsWith(string suffix) Returns true if this string object ends with string suffix, and false
otherwise. Example:
csharp> "-123".EndsWith("-");
false
csharp> "downstairs".EndsWith("airs");
true
csharp> "downstairs".EndsWith("air");
false

6.6.1 Count Repetitions in a String Exercise

Write a program test_count_rep.cs, with a Main testing method, that tests a function with the following head-
ing:
// Return the number of separate repetitions of target in s.
static int CountRep(string s, string target)

For example here is what CountRep( "Mississippi", target) would return with various values for
target:
"i": 4
"is": 2
"sss": 0
Assume each repetition is completely separate, so CountRep("Wheee!", "ee") returns 1. The last two e’s do
not count, since the middle e is already used in the match of the first two e’s.

90 Chapter 6. While Loops

 Introduction to Computer Science in C#, Release 2.0

6.6.2 Safe PromptInt and PromptDouble Exercise

Save the example safe_number_input_stub/safe_number_input.cs in a project of your own.

The idea is to write safe versions of the utility functions PromptInt and PromptDouble (which can then be used in
further places like PromptIntInRange).
Be sure you are familiar with Safe Whole Number Input Exercise, and the development of its InputWhole function.
A legal whole number string consists entirely of digits. We have already written example IsDigits to identify a
string for a whole number.
The improvements to PromptInt and PromptDouble are very similar and straightforward if you have developed the
two main Boolean support functions, IsIntString and IsDecimalString respectively.
A complicating issue with integer and decimal strings is that they may include parts other than digits. An integer may
start with a minus sign. A decimal number can also contain a decimal point in an appropriate place. The suggestion
is to confirm that these other characters appear in legal places, remove them, and see that what is left is digits. The
recently introduced string methods should help....
Develop the functions in order and test after each one: write IsIntString, revise PromptInt, write
IsDecimalString, and revise PromptDouble.
Be sure to test carefully. Not only confirm that all appropriate strings return true: Also be sure to test that you return
false" for all sorts of bad strings.
Hopefully you learned something from writing the earlier PromptWhole. Probably it is not worth keeping in our
utility library any longer, since we have the more general and safe PromptInt, and we can restrict to many ranges with
PromptIntInRange.
We will arrange for these functions to be a library class later. For now just develop and test them in this one class.

6.7 User Input: UI

With the exercises in the last section, we have all we need for a much improved User Input library class. This will be the
class UI. We have the original PromptLine, improved PromptInt, PromptDouble, PromptIntInRange,
PromptDoubleInRange, Agree, and assorted supporting functions. The only changes to individual methods
were to make sure that the static methods are public.
Henceforth we will be using UI in place of UIF. In fact all the places UIF was used before could be replaced by UI:
It includes all the functionality of UIF.
You can look at the code all together in ui/ui.cs.
The functions are still not perfect: It is possible to blow things up by entering too long an integer. Though that
could be addressed with our present technology, it probably makes sense to wait for an introduction of Exception
handling syntax to really make things bulletproof. Meanwhile you have good examples of interactive loops and string
manipulation.

6.8 Greatest Common Divisor

6.8.1 Euclid’s Algorithm

The greatest common divisor of two non-zero integers is a great example to illustrate the power of loops. Everyone
learns about the concept of a greatest common divisor when faced with a fraction that is not in reduced form.

6.7. User Input: UI 91

Introduction to Computer Science in C#, Release 2.0

Consider the fraction 24 , which is the same as 12 . This fraction can be reduced, because the numerator and denominator
both have greatest common factor of 2. That is, 42 = 2·2 1·2
. So the factor of 2 can be canceled from both the numerator
and the denominator.
Euclid (the mathematician from classic times and author of Elements) is credited with having come up with a clever
algorithm for how to compute the greatest common divisor efficiently. It is written as follows, where 𝑎 mod 𝑏 means
a % b in C#.
𝑔𝑐𝑑(𝑎, 𝑏) = 𝑔𝑐𝑑(𝑏, 𝑎 mod 𝑏)
𝑔𝑐𝑑(𝑎, 0) = 𝑎
It is common in mathematics to list functions as one or more cases. The way you read this is as follows:
• In general, the greatest common divisor of a and b is the same as computing the greatest common divisor of b
and the remainder of a divided by b.
• In the case where b is zero, the result is a. This makes sense because a divides itself and 0.
To gain some appreciation of how the definition always allows you to compute the greatest common divisor, it is
worthwhile to try it out for a couple of numbers where you know the greatest common divisor. For example, we
already know that the greatest common divisor of 10 and 15 is 5. Let’s use Euclid’s method to verify this:
• 𝑔𝑐𝑑(10, 15) = 𝑔𝑐𝑑(15, 10 mod 15) = 𝑔𝑐𝑑(15, 10)
• 𝑔𝑐𝑑(15, 10) = 𝑔𝑐𝑑(10, 15 mod 10) = 𝑔𝑐𝑑(10, 5)
• 𝑔𝑐𝑑(10, 5) = 𝑔𝑐𝑑(5, 10 mod 5) = 𝑔𝑐𝑑(5, 0)
• 𝑔𝑐𝑑(5, 0) = 5
Notice that in the example above, the first number (10) was smaller than the second (15), and the first transformation
just swapped the numbers, so the larger number was first. Thereafter the first number is always larger.
In other words, Euclid’s method is smart enough to work for 10 and 15 and 15 and 10. And it must. After all, the
greatest common divisor of these two numbers is always 5 as the order doesn’t matter.

6.8.2 GCD “Brute Force” Method

Now that we’ve gotten the preliminaries out of the way and have a basic mathematical explanation for how to calculate
the greatest common divisor, we’ll take a look at how to translate this into code using the machinery of while loops
that you’ve recently learned.
The way GCD is formulated above is, indeed, the most clever way to calculate the greatest common divisor. Yet the
way we learn about the greatest common divisor in elementary school (at least at first) is to learn how to factor the
numbers a and b, often in a brute force way. So for example, when calculating the greatest common divisor of 10 and
15, we can immediately see it, because we know that both of these numbers are divisible by 5 (e.g. 5 * 2 = 10 and 5 *
3 = 15). So the greatest common divisor is 5.
But if we had something more tricky to do like 810 and 729, we might have to think a bit more.
Before we learn to find the factors of numbers, we will often just “try” numbers until we get the greatest common
divisor. This sort of trial process can take place in a loop, where we start at 1 and end at min(a, b). Why the minimum?
We know that none of the values after the minimum can divide both a and b (in integer division), because no larger
number can divide a smaller positive number. The smaller number would be the (non-zero) remainder.
Now take a look at a basic version of GCD:
This code works as follows:
• We begin by finding Math.Min(a, b). This is how to compute the minimum of any two values in C#.
Technically, we don’t need to use the minimum of a and b, but there is no point in doing any more work than
necessary.

92 Chapter 6. While Loops

 Introduction to Computer Science in C#, Release 2.0

• We’ll use the variable i as the loop index, starting at 1.

• The variable gcd will hold the largest currently known common divisor. We start with 1, which divides any
integer, and we will look for a higher value that also divides a and b.
• The line while (i <= n) is used to indicate that we are iterating the values of i until the minimum of a
and b (computed earlier) is reached.
• The line if (a % i == 0 && b % i == 0) is used to check whether we have found a new value that
replaces our previous candidate for the GCD. A value can only be a candidate for the GCD if it divides a and b
without a remainder. The modulus operator % is our way of determining whether there is a remainder from the
division operation a / i or b / i.
• The line i++ is our way of going to the next value of i to be tested as the new GCD.
• When this loop terminates, the greatest common divisor has been found.
So this gives you a relatively straightforward way of calculating the greatest common divisor. While simple, it is not
necessarily the most efficient way of determining the GCD. If you think about what is going on, this loop could run a
significant number of times. For example, if you were calculating the GCD two very large numbers, say, one billion
(1,000,000,000) and two billion (2,000,000,000) it is painfully evident that you would consider a large number of
values (a billion, in fact) before obtaining the candidate GCD, which we know is 1,000,000,000.

Brute-Force GCD Exercise

The code above goes though all integers 2 through min(a, b). That is not generally necessary when the GCD
is greater than 1, even with a brute-force mindset. Write a g_c_d_basic_faster.cs to do this with a slightly
different GreatestCommonDivisor function.

6.8.3 GCD Subtraction Method

The subtraction method (also attributable to Euclid) to compute the Greatest Common Divisor works as follows:
• Based on the mathematical definition in the previous section, the greatest common divisor algorithm saves a
step when we already have a and b in the right order.
• The right order means that 𝑎 > 𝑏. As we noted earlier, the cleverness of the mathematical definition is that a
and b are swapped as the first step to ensure that 𝑎 > 𝑏, after which we can repeatedly divide to get the GCD.
• Division, of course, is a form of repetitive subtraction, so the way to divide by b is to repeatedly subtract it (from
a) until a is no longer greater than b.
• The subtraction method basically makes no attempt to put a and b in the right order. Instead, we just write
similar loops to allow for the possibility of either order.
• A simple check must be performed to ensure that the approach of repeated subtraction actually resulted in the
GCD. This will happen if a and b bump into one another, thereby meaning that we have computed the GCD.
A look at the source code more or less follows the above explanation.
Let’s start by looking at the inner loop at line 5, while (a > b). In this loop, we are repeatedly subtracting b from
a, which we know we can do, because a started out as being larger than b. At the end of loop a is reduced to either
1. b, in which case b exactly divided the earlier a, and b is the GCD, or
2. a number less than b, namely 𝑎 mod 𝑏 (or in C# terms a % b), and the process continues....
The loop on line 9 is similar to the loop in line 5. For the same reasons as we already explained, b ends up equal to a,
which is the GCD, or b ends up as 𝑏 mod 𝑎.

6.8. Greatest Common Divisor 93

Introduction to Computer Science in C#, Release 2.0

As discussed above, if a and b end up as the same number, that is the GCD. On the other hand, the first GCD algorithm
example showed how remainders may need to be to be calculated over and over. The outer loop in this version keeps
this up until a and b are reduced to equal values. At this point the inner loops would make no further changes, and the
common value is the GCD.
As an exercise to the reader, you may want to consider adding some Console.WriteLine() statements to print
the values of a and b within each loop, and after both loops have executed. It will allow you to see in visual terms
how this method does its work.

6.8.4 GCD Remainder Loop

There are several ways to code the shorter Euclidean algorithm at the beginning of this GCD section. It is a repetitive
pattern, and a loop can be used. There are two parameters, a and b, to the gcd, and they can be successively changed,
suggesting a loop. What is the continuation condition? You stop when b is 0, so you continue while b != 0. The
parameters a and b need to be replaced by b and a % b. One extra variable needs to be introduced to make this
double change work. The simplest is to introduce a variable r for the remainder. Check and see for yourself that you
need an extra variable like r. This code is from g_c_d_remainder_loop/g_c_d_remainder_loop.cs:

6.8.5 Preview: Recursive GCD

The first statement of Euclid’s algorithm said (in C#) when

gcd(a, b) = gcd(b, a % b)

It is saying the result of the function with one set of parameters is equal to calling the function with another set of
parameters. If we put this into a C# function definition, ti would mean the instructions for the function say to call
itself. This is a broadly useful technique called recursion, where a function calls itself inside its definition. We don’t
expect you to master this technique immediately but do feel that it is important you at least hear about it and see its
tremendous power:
• Recalling our earlier definition, the case 𝑔𝑐𝑑(𝑎, 0) = 𝑎 is handled by lines 3-5.
• And the case 𝑔𝑐𝑑(𝑎, 𝑏) = 𝑔𝑐𝑑(𝑏, 𝑎 mod 𝑏) is handled by line 10.
• Lines 4 and 7-9 exist to show you all of the steps that Euclid’s algorithm takes to compute the greatest common
divisor.
The recursive version of the gcd function refers to itself by calling itself. Though this seems circular, you can see
from the examples that it works very well. The important point is that the calls to the same function are not completely
the same: Successive calls have smaller second numbers, and the second number eventually reaches 0, and in that case
there is a direct final answer. Hence the function is not really circular.
This recursive version is a much more direct translation of the original mathematical algorithm than the looping
version.
The general idea of recursion is for a function to call itself with simpler parameters, until a simple enough place is
reached, where the answer cam be directly calculated.

6.9 Do-While Loops

Suppose you want the user to enter three integers for sides of a right triangle. If they do not make a right triangle, say
so and make the user try again. x One way to look at the while statement rubric is:

94 Chapter 6. While Loops

 Introduction to Computer Science in C#, Release 2.0

set data for condition

while (condition) {
accomplish something
set data for condition
}

As we have pointed out before this involves setting data in two places. With the triangle problem, three pieces for
data need to be entered, and the condition to test is fairly simple. (In any case the condition could be calculated in a
function.)
A do-while loop will help here. It tests the condition at the end of the loop, so there is no need to gather data before
the loop:
int a, b, c;
do {
Console.WriteLine("Think of integer sides for a right triangle.");
a = UI.PromptInt("Enter integer leg: ");
b = UI.PromptInt("Enter another integer leg: ");
c = UI.PromptInt("Enter integer hypotenuse: ");
if (a*a + b*b != c*c) {
Console.WriteLine("Not a right triangle: Try again!");
}
} while (a*a + b*b != c*c);

The general form of a do-while statement is

do {
statement(s)
} while ( continuationCondition );
Here the block of statement(s) is always executed at least once, but it continues to be executed in a loop only so long
as the condition tested after the loop body is true.

Note: A do-while loop is the one place where you do want a semicolon right after a condition, unlike the places
mentioned in Dangerous Semicolon. At least if you omit it here you are likely to get a compiler error rather than a
difficult logical bug.

A do-while loop, like the example above, can accomplish exactly the same thing as the while loop rubric at the
beginning of this section. It has the general form:
do {
set data for condition
if (condition) {
accomplish something
}
} while (condition);

It only sets the data to be tested once. (The trade-off is that the condition is tested twice.)

6.10 Number Guessing Game Lab

Objectives:
• Work with functions
• Work with interactive while loops

6.10. Number Guessing Game Lab 95

Introduction to Computer Science in C#, Release 2.0

• Use decisions
• Introduce random values
This lab is inspired by a famous children’s game known as the number-guessing game. We suppose two people are
playing.
The rules are:
• Person A chooses a positive integer less than N and keeps it in his or her head.
• Person B makes repeated guesses to determine the number. Person A must indicate whether the guess is higher
or lower.
• Person A must tell the truth.
So as an example:
• George and Andy play the game.
• George chooses a positive number less than 100 (29) and puts it in his head.
• Andy guesses 50. George says “Lower”. Andy now knows that 1 ≤ 𝑛𝑢𝑚𝑏𝑒𝑟 < 50.
• Andy guesses 25. George says “Higher”. Andy now knows that 26 ≤ 𝑛𝑢𝑚𝑏𝑒𝑟 < 50.
• Andy guesses 30. George says “Lower”. Andy now knows that the 26 ≤ 𝑛𝑢𝑚𝑏𝑒𝑟 < 30.
• Andy starts thinking that he is close to knowing the correct answer. He decides to guess 29. Andy guesses the
correct number. So George says, “Good job! You win.”
We are going to elaborate this game in small steps. You might save the intermediate versions under new names.
The computer code for the game is going to be acting like Player A.

6.10.1 Part 1: No Hints; Fixed Secret Number

If using Xamarin Studio, create a new project in a solution in which you already have added the ui library project. Add
the ui project as a reference for the lab project. Make sure your program has namespace IntroCS; to match the
ui project.
You are going to play a game, and later may repeat it, so put the code for playing the number game in a function called
Game:
static void Game()

For now your Main function can just call Game().

In Game:
1. For the simplest versions, which help testing, have the program assign a specific secret number (like 29), and
call it secret. Admittedly, this is not much fun for the player the second time!
2. Prompt the player for a guess. Use UI.PromptInt. Every time the player guesses wrong, print “Wrong!”. A
later version will give clues. Keep prompting for another number until the player guesses correctly. (Since you,
the programmer, knows the secret number, this need not go on forever.)
3. When the player guesses the right number, print “Correct! You win!”
Sample play could look like:
Guess the number: 55
Wrong!
Guess the number: 12

96 Chapter 6. While Loops

 Introduction to Computer Science in C#, Release 2.0

Wrong!
Guess the number: 29
Good job! You win!
You could also make the game stop immediately, (since you know the secret number):
Guess the number: 29
Good job! You win!

6.10.2 Part 2: Add Hints

In Game: Instead of just printing “Wrong!” when the player is incorrect, print “Lower!” or “Higher!” as appropriate.
For example:
Guess the number: 55
Lower!
Guess the number: 12
Higher!
Guess the number: 25
Higher!
Guess the number: 29
Good job! You win!

6.10.3 Part 3: Add a Random Secret Number

In Game, make the following alterations and additions:

1. For now set an int variable big to 100. We will make sure the secret number is less than big.
2. Have the program print “In this game you guess a positive number less than 100.” For future use it is best if you
have the printing statement reference the variable big, rather than the literal 100.
1. Thus far the secret number was fixed in the program. Now we are going to let it vary, by having the game
generate a random number. For your convenience, we are going to give you the C# code to compute the random
number. Assuming we want a secret number so 1 ≤ 𝑠𝑒𝑐𝑟𝑒𝑡 < 𝑏𝑖𝑔, we can use the code:
Random r = new Random();
int secret = r.Next(1, big);

In case you are wondering, we are creating a new object of the class Random which serves as the random
number generator. We’ll cover this in more detail when we get to the Classes and Object-Oriented Programming
chapter. Here is some illustration in csharp. Your answers will not be the same!
csharp> Random r = new Random();
csharp> r.Next(1, 100);
55
csharp> r.Next(1, 100);
31
csharp> r.Next(1, 100);
79
csharp> r.Next(2, 5);
3
csharp> r.Next(2, 5);
4
csharp> r.Next(2, 5);
3

6.10. Number Guessing Game Lab 97

Introduction to Computer Science in C#, Release 2.0

csharp> r.Next(2, 5);

2

In general the minimum possible value of the number returned by r.Next is the first parameter, and the value
returned is always less than the second parameter, never equal.
You can see that r.Next() is smart enough to give what appears to be a randomly chosen number every time.
Example (where secret ended up as 68):
Guess a number less than 100!
Guess the number: 60
Higher!
Guess the number: 72
Lower!
Guess the number: 66
Lower!
Guess the number: 68
Good job! You win!
For debugging purposes, you might want to have secret be printed out right away. (Eliminate that part when
everything works!)

6.10.4 Part 4: Let the Player Set the Range of Values

In Game: Instead of setting declaring big and automatically initializing it to 100, make big be a parameter, so the
heading looks like:
static void Game(int big)

In Main:
1. Prompt the player for the limit on the secret number. An exchange might look like:
Enter a secret number bound: 10
2. Pass the value given by the player to the Game function.
Hence the program might start with:
Enter a secret number bound: 10
Guess a number less than 10!
Guess the number: 5
Higher!
Guess the number: 7
Lower!
Guess the number: 6
Good job! You win!

6.10.5 Part 5: Count the Guesses

In Game: When the player finally wins, print the number of guesses the player made. For example, for the game
sequence shown above, the last line would become:
Good job! You win on guess 3!

98 Chapter 6. While Loops

 Introduction to Computer Science in C#, Release 2.0

You need to keep a count, adding 1 with each guess.

6.10.6 Possible Extra Credit Improvements or Variations

Should you finish everything early, try the following:

1. (20% extra credit) In Main:
Use an outer while loop to allow the game to be played repeatedly. Change the prompt for the bound in Main
to:
Enter a secret number bound (or 0 to quit):
Continue to play games until the player enters 0 for the bound.
2. (40% extra credit) In Main prompt users to see if they want to guess numbers or reverse roles and choose the
secret number. In the first case, just call the existing Game function. In the second case you need a new function,
where the user is the one who knows the secret number and the computer guesses numbers until the answer is
obtained. Write and use a new function
static void GameReversed(int big)

Pass it the parameter big, still set in Main. The new GameReversed will tell the user to put a number
in his/her head, and press return to continue. (You can throw away the string entered - this is just to cause a
pause.) Then the computer guesses. For simplicity let the human enter “L” for lower, “H” for higher, and “E”
for equal (when the computer wins). As you saw in the initial example with George and Andy, each hint reduces
the range of the possible secret numbers. Have the computer guess a random number in the exact range that
remains possible.
To do this you must note the asymmetry of the parameters for the method Next: suppose n = r.Next(low,
higher), then
𝑙𝑜𝑤 ≤ 𝑛𝑢𝑚𝑏𝑒𝑟 < ℎ𝑖𝑔ℎ𝑒𝑟
The first parameter may be returned, but second parameter is never returned.
You will need two parameters low and higher that keep bracketing the allowed range. The simplest thing is
to set them so they will be the parameters for the following call to Next.
That would mean initially low is 1 and higher is equal to big. With each hint you adjust one or the other of
low and higher so they get closer together. The game ends after the human enters “E”.
Have the computer complain that the human is cheating (and stop the game) if the computer guesses the only
possible value, and the human does not respond with “E”.

6.10. Number Guessing Game Lab 99

Introduction to Computer Science in C#, Release 2.0

100 Chapter 6. While Loops

 CHAPTER

SEVEN

FOREACH LOOPS

7.1 foreach Syntax

This sections on foreach loops and the later For Loops introduce new looping statements. Neither is absolutely
necessary: You could do all the same things with while loops, but there are many situations where foreach loops
and for loops are more convenient and easier to read.
A foreach statement only works with an object that holds a sequence or collection. We will see many more kinds
of sequences later. For now we can illustrate with a string, which is a sequence of characters.
We have already processed strings a character at a time, with while loops. We took advantage of the fact that strings
could be indexed, and our while loops directly controlled the sequence of indices, and then we could look up the
character at each index of a given string s:
int i = 0;
while (i < s.Length) {
use value of s[i]...
i++;
}

Examples have been in While-Statements with Sequences, like

In this example we really only care about the characters, not the indices. Managing the indices is just a way to get at
the underlying sequence of characters.
A conceptually simpler view is just:
for each character in s
use the value of the character

To use “the character” in C#, we must be able to refer to it. We might name the current character ch. The following
is a variant of OneCharPerLine with a foreach loop:
static void OneCharPerLine(string s)
{
foreach (char ch in s) {
Console.WriteLine(ch);
}
}

That is all you need! The foreach heading feeds us one character from s each time through, using the name ch
to refer to it. Of course any new variable name must be declared in C#, so ch is preceded in the heading by its type,
char. Then we can use ch inside the body of the loop. Advancing to the next element in the sequence is automatic
in the next time through the loop. No i++ to remember; no possibility of an infinite loop!
The general syntax of a foreach loop is

101
Introduction to Computer Science in C#, Release 2.0

foreach ( type itemName in sequence ) {

statement(s) using itemName
}
Here is a version of IsDigits:
static Boolean IsDigits(string s)
{
foreach (char ch in s) {
if (ch < ’0’ || ch > ’9’) {
return false;
}
}
return (s.Length > 0);
}

See the advantages of foreach in these examples:

• They are more concise than the indexing versions.
• They keep the emphasis on the characters, not the secondary indices.
• The foreach heading emphasizes that a particular sequence is being processed.

Warning: If you have explicit need to refer to the indices of the items in the sequence, then a foreach statement
does not work.

Of course you can refer to the indices of items in sequence with a flexible while loop, or see For Loops, coming
soon....

7.2 foreach Examples

In IsDigits we use the underlying int Unicode value of the characters in comparisons. When printing, you cannot see
this code directly, since the char type prints as characters! To see the underlying code value for a character, ch, it
can be cast to an int: (int)ch
We can easily write a loop to print the unicode value of each character in a string, s. We do not need indices here, so
a foreach loop is appropriate:
foreach (char ch in s) {
Console.WriteLine("Unicode for {0} is {1}.", ch, (int)ch);
}

Try this in csharp.

We will have many more examples after we introduce more kinds of sequences.

102 Chapter 7. Foreach Loops

 CHAPTER

EIGHT

FOR LOOPS

8.1 For-Statement Syntax

We now introduce the last kind of loop syntax: for loops.

A for loop is an example of syntactic sugar: syntax that can simplify things for the programmer, but can be immedi-
ately translated into an equivalent syntax by the compiler. For example:
for (i = 2; i <= n; i++) {
sum = sum + i;
}

is exactly equivalent to this code similar to part of SumToN:

i= 2;
while (i <= n) {
sum = sum + i;
i++;
}

More generally:
for ( initialization ; condition ; update ) {
statement(s)
}
translates to
initialization ;
while ( condition ) {
statement(s)
update
}
In the example above, initialization is i=2, condition is i <= n, and update is i++.
Why bother with this rearrangement? It is a matter of taste, but the heading:
for (i = 2; i <= n; i++) {

puts all the information about the variable controlling the loop into one place at the top, which may help quickly
visualize the overall sequence in the loop. If you use this format, and get used to the three parts you are less likely to
forget the i++ than when it comes tacked on to the end of a while loop body, after all the specific things you were
trying to accomplish.

103
Introduction to Computer Science in C#, Release 2.0

Although the for loop syntax is very general, a strongly recommended convention is to only use a for statement
when all the control of variables determining loop repetition are in the heading.
For example if a for loop uses i in the heading, i can have a value assigned or reassigned in the heading, but should
not have its value modified anywhere inside the loop body. If you want more complicated behavior, use a while loop.
A for loop can also include variable declaration in the initialization, as in:
for (int i = 2; i <= n; i++) {
sum = sum + i;
}

This is close, but not quite equivalent to:

int i = 2;
while (i <= n) {
sum = sum + i;
i++;
}

Variables declared in a for loop heading are local to the for loop heading and body. The variable i declared before
the while statement above is still defined after the while loop.
The two semicolons are always needed in the for heading, but any of the parts they normally separate may be omitted.
If the condition part is omitted, the condition is interpreted as always true, leading to an infinite loop, that can only
terminate due to a return or break statement in the body.
Other variations
As in a regular local variable declaration, there may be several variables of the same type initialized at the beginning of
a for loop heading, separated by commas. Also, at the end of the for loop heading, the update portion may include
more than one expression, separated by commas. For example:
for (int i = 0, j = 10; i < j; i = i+2, j++) {
Console.WriteLine("{0} {1}", i, j);
}

The comma separated lists in a for statement heading are mentioned here for completeness. Later we will find a
situation where this is actually useful.

8.1.1 Break and Continue

This section concerns special break and continue statements that can only occur inside a loop (any kind: while, for
or foreach). The syntax is convenient in various circumstances, but not necessary. You are free to use it, but for this
course it is an optional extra:
You can already stop a loop in the middle with an if statement that leads to a choice with a return statement. Of
course that forces you to completely leave the current function. If you only want to break out of the innermost current
loop, but not out of the whole function, use a break statement:
break;
in place of return. Execution continues after the end of the whole innermost currently running loop statement. The
break and continue statements only make practical sense inside of an if statement that is inside the loop.
Examples, assuming target already has a string value and a is an array of strings:
bool found = false;
for (int i = 0; i < a.Length; i++) {
if (a[i] == target) {
found = true;

104 Chapter 8. For Loops

 Introduction to Computer Science in C#, Release 2.0

break;
}
}
if (found) {
Console.WriteLine("Target found at index " + i);
} else {
Console.WriteLine("Target not found");
}

When an element is reached that matches target, execution goes on past the loop with if (found) ....
An alternate implementation with a compound condition in the heading and no break is:
bool found = false;
for (int i = 0; i < a.Length && !found; i++) {
if (a[i] == target) {
found = true;
}
}
if (found) {
Console.WriteLine("Target found at index " + i);
} else {
Console.WriteLine("Target not found");
}

With a foreach loop, which has no explicit continuation condition, the break would be more clearly useful. Here
is a variant if you do not care about the specific location of the target:
bool found = false;
foreach (string s in a) {
if (s == target) {
found = true;
break;
}
}
if (found) {
Console.WriteLine("Target found");
} else {
Console.WriteLine("Target not found");
}

Using break statements is a matter of taste. There is some advantage in reading and following a loop that has only
one exit criteria, which is easily visible in the heading. One the other hand, in many situations, using a break statement
makes the code much less verbose, and hence easier to follow. If you are reading through the loop, it may be clearer
to have an immediate action where it is certain that the loop should terminate.
All the modifiers about innermost loop are important in a situation like the following:
for (....) {
for (....) {
...
if (...) {
...
break;
}
...
}
}

The break statement is in the inner loop. If it is reached, the inner loop ends, but the inner loop is just a single statement

8.1. For-Statement Syntax 105

Introduction to Computer Science in C#, Release 2.0

inside the outer loop, and the outer loop continues. If the outer loop continuation condition remains true, the inner
loop will be executed again, and the break may or may not be reached that time, so the inner loop may or may not
terminate normally....
For completeness we mention the much less used continue statement:
continue;
It does not break out of the whole loop: It just skips the rest of the body of the innermost current loop, this time through
the loop. In the simplest situations a continue statement just avoids an extra else clause. It can considerable
shorten code if the test is inside complicated, deeply nested if statements.

8.2 Examples With for Statements

Thus far all of our for loops have used a sequence of successive integers. Suppose you want to print the first n
multiples of k, like the first 5 multiples of 3: 3, 6, 9, 12, 15. This could be handled by generating a sequence i = 1
through n, and multiply each i by k:
for (int i = 1; i <= n; i++) {
Console.WriteLine(i*k);
}

Another approach is to note that the numbers you want to print advance in a regular fashion, too, but with an increment
3 in the example above, or k in general:
for (int i = k; i <= n*k; i = i+k) {
Console.WriteLine(i);
}

The
i = i + k;

is a common pattern, less common than incrementing by one, but still very common. C# and many other languages
allow a shorter version:
i += k;

This means to increment the variable i by k.

Warning: Be careful: the += must be in that order, with no space between. Unfortunately the reverse order:

i =+ k;

is also legal, and just assigns the value of k to i.

Most C# binary operations have a similar variation. For instance if op is +, -, *, / or %,

variable op= expression
means the same as
variable = variable op expression
For example
x *= 5;

is the same as

106 Chapter 8. For Loops

 Introduction to Computer Science in C#, Release 2.0

x = x * 5;

8.2.1 Tables

Reports commonly include tables, often with successive lines generated by a consistent formula. As a simple first
table can show the square, cube, and square root of numbers 1 through 10. The Math class has a function Sqrt, so we
take the square root with Math.Sqrt function. The formula is consistent, so we can loop easily:
for ( int n = 1; n <= 10; n++) {
Console.WriteLine("{0} {1} {2} {3}", n, n*n, n*n*n, Math.Sqrt(n));
}

The numbers will be there, but it is not pretty:

1 1 1 1
2 4 8 1.4142135623731
3 9 27 1.73205080756888
4 16 64 2
5 25 125 2.23606797749979
6 36 216 2.44948974278318
7 49 343 2.64575131106459
8 64 512 2.82842712474619
9 81 729 3
10 100 1000 3.16227766016838

First we might not need all those digits in the square root approximations. We can replace {3} by {3:F4} to just
show 4 decimal places.
There are not nice columns lining up. We can adjust the spacing to make nice columns by using a further formatting
option. The longest entries are all in the last row, where they take up, 2, 3, 4, and 6 columns (for 3.1623). Change the
format string:
for ( int n = 1; n <= 10; n++) {
Console.WriteLine("{0,2} {1,3} {2,4} {3,6:F4}",
n, n*n, n*n*n, Math.Sqrt(n));
}

and we generate the neater:

1 1 1 1.0000
2 4 8 1.4142
3 9 27 1.7321
4 16 64 2.0000
5 25 125 2.2361
6 36 216 2.4495
7 49 343 2.6458
8 64 512 2.8284
9 81 729 3.0000
10 100 1000 3.1623

We are using two new formatting forms:

{index,fieldWidth} and
{index,fieldWidth:F#}
where index, fieldWidth, and # are replaces by specific integers. The new part with the comma (not colon) and
fieldWidth, sets the minimum number of columns used for the substituted string, padding with blanks as needed.

8.2. Examples With for Statements 107

Introduction to Computer Science in C#, Release 2.0

Warning: There is a special language for the characters between the braces in a format string. The rules are
different than in regular C# code, where comma and colon are symbols, and the parser allows optional whitespace
around them. This is not the case inside the braces of a format string: There cannot be a space after the colon or
before the comma. Some blanks are legal; some blanks lead to exceptions being thrown, and other positions for
blanks just silently give the wrong format.
The safest approach for a programmer is just to have no blanks between the braces in a format string.

If the string to be inserted is wider than the fieldWidth, then the whole string is inserted, ignoring the fieldWidth.
Example:
string s = "stuff";
Console.WriteLine("123456789");
Console.WriteLine("{0,9}\n{0,7}\n{0,5}\n{0,3}", s);

generates:
123456789
stuff
stuff
stuff
stuff

filling 9, 7, and then 5 columns, by padding with 4, 2, and 0 blanks. The last line sticks out past the proposed 3-column
fieldWidth.
One more thing to add to our power table is a heading. We might want:
n square cube root

To make the data line up with the heading titles, we can expand the columns, with code in example
power_table/power_table.cs:
generating:
n square cube root
1 1 1 1.0000
2 4 8 1.4142
3 9 27 1.7321
4 16 64 2.0000
5 25 125 2.2361
6 36 216 2.4495
7 49 343 2.6458
8 64 512 2.8284
9 81 729 3.0000
10 100 1000 3.1623

Note how we make sure the columns are consistent in the heading and further rows: We used a format string for the
headings with the same field widths as in the body of the table. A separate variation: We also reduced the length of
the format string by putting all the substitution expressions in braces right beside each other, and generate the space
between columns with a larger field width.
Left Justification: Though our examples have always right justified in a field (padding on the left), for completeness
we note this alternative: A minus sign in front of the fieldWidth places the result left justified (padded on the right).
For example:
string s = "stuff";
Console.WriteLine("1234567890");
Console.WriteLine("{0,-9}|\n{0,-7}|\n{0,-5}|\n{0,-3}|", s);

108 Chapter 8. For Loops

 Introduction to Computer Science in C#, Release 2.0

prints:
1234567890
stuff |
stuff |
stuff|
stuff|

where the ‘|’ appears after any blank padding in each line.

8.2.2 ASCII Codes

Here is a reverse lookup from the Numeric Code of String Characters: Find the characters for a list of numeric codes.
Just as we can cast a char to an int, we can cast an int 0-127 to a char.
The Unicode used by C# is an extension of the ASCII codes corresponding to the characters on a US keyboard.
The codes were originally used to drive printers, and the first 32 codes are non-printable instructions to the printer.
Characters 32 - 126 yield the 95 characters on a standard US keyboard.
A loop to print each code followed by a space and the corresponding printable character would be:
for (int i = 32; i < 127; i++) {
Console.WriteLine("{0,3} {1}", i, (char)i);
}

To make all the character line up we added a field width 3 for the code column.
If you run this in csharp, the first line printed does not appear to have a character: That is the blank character. All the
other characters are visible.
Let us make a more concise table, putting 8 entries per line. We can print successive parts using Write instead of
WriteLine, but we still need to advance to the next line after every 8th entry, for codes 39, 47, 55, .... Since they
are 8 apart, their remainder when divided by 8 is always the same:
7 = 39 % 8 = 47 % 8 = 55 % 8 = ....
We can add a newline after each of these is printed. This requires a test:
for (int i = 32; i < 127; i++) {
Console.Write("{0,3} {1} ", i, (char)i);
if (i % 8 == 7) {
Console.WriteLine();
}
}

Recall that Console.WriteLine() with no parameters only advances to the next line.
Paste that whole code at once into csharp to see the result.
The next csharp> prompt appears right after 126 ~. There is no eighth entry on the last line, and hence no advance
to the next line. A program printing this table should include an extra Console.WriteLine() after the loop.

8.2.3 Modular Multiplication Table

We have introduced the remainder operator % and mentioned have the corresponding mathematical term is “mod”. We
can extend that to the idea of modular arithmetic systems. For example, if we only look at remainders mod 7, we can
just consider numbers 0, 1, 2, 3, 4, 5, and 6. We can do multiplication and addition and take remainders mod 7 to get
answers in the same range. For example 3 * 5 mod 7 is (3 * 5) % 7 in C#, which is 1. As we look more at this
system, we will observe and explain more properties.

8.2. Examples With for Statements 109

Introduction to Computer Science in C#, Release 2.0

The next example is to make a table of multiplication, mod 7, and later generalize.
Tables generally have row and column labels. We can aim for something like:

* | 0 1 2 3 4 5 6
-----------------
0 | 0 0 0 0 0 0 0
1 | 0 1 2 3 4 5 6
2 | 0 2 4 6 1 3 5
3 | 0 3 6 2 5 1 4
4 | 0 4 1 5 2 6 3
5 | 0 5 3 1 6 4 2
6 | 0 6 5 4 3 2 1

The border labels make the table much more readable, but let us start simpler, with just the modular multiplications:
0 0 0 0 0 0 0
0 1 2 3 4 5 6
0 2 4 6 1 3 5
0 3 6 2 5 1 4
0 4 1 5 2 6 3
0 5 3 1 6 4 2
0 6 5 4 3 2 1

This is more complicated in some respects than our previous table, so start slow, with some pseudocode. We need a
row for each number 0-6, and so a for loop suggests itself:
for (int r = 0; r < 7; r++) {
print row
}

Each individual row also involves a repeated pattern: calculate for the next number. We can name the second number
c for column. The next revision replaces “print row” by a loop: a nested loop, inside the loop for separate rows:
for (int r = 0; r < 7; r++) {
for (int c = 0; c < 7; c++) {
print modular multiple on same line
}
}

and the modular multiplication is just regular multiplication followed by taking the remainder mod 7, so you might
come up with the C# code:
for (int r = 0; r < 7; r++) {
for (int c = 0; c < 7; c++) {
int modProd = (r*c) % 7;
Console.Write(modProd + " ");
}
}

You can test this in csharp, and see it is not quite right! chopped-off output starts:
0 0 0 0 0 0 0 0 1 2 3 4 5 6 0 2 4 6 1 3 5 0 3 6 2 5 1 4 0...

Though we want each entry in a row on the same line, we need to go down to the next line at the end of each line!
Where do we put in the newline in the code? A line is all the modular products by r, followed by one newline. All the
modular products for a row are printed in the inner for loop. We want to advance after that, so the newline must be
inserted outside the inner loop. On the other hand we do want it done for each row, so it must be inside the outer loop:

110 Chapter 8. For Loops

 Introduction to Computer Science in C#, Release 2.0

1 for (int r = 0; r < 7; r++) {

2 for (int c = 0; c < 7; c++) {
3 int modProd = (r*c) % 7;
4 Console.Write(modProd + " ");
5 }
6 Console.WriteLine();
7 }

You can copy and test that code in csharp, and it works!
It is important to be able to play computer on nested loops and follow execution, statement by statement. Look more
closely at the code above, noting the added line numbers. General sequencing orders apply: The basic pattern is
sequential: Complete one statement before going on to the next. Inside the execution of a looping statement, there are
extra rules, for testing and going through the whole loop body sequentially. Most new students can get successfully to
line 4:
line r c modProd comment
1 0 - - initialize outer loop
2 0 0 - initialize inner loop
3 0 0 0
4 0 0 0 Write 0
After reaching the bottom of the loop, where do you go? You finish the innermost statement that you are in. You are in
the inner loop, so the next line is the inner loop heading where you increment c and continue with the loop since 1 <
7. This inner loop continues until you reach the bottom of the inner loop, line 4, with c = 6, and return to the heading,
line 2, and the test fails, finishing the inner row loop:
line r c modProd comment
1 0 - - initialize outer loop
2 0 0 - 0 < 7, enter loop body
3 0 0 0 (0*0)%7
4 0 0 0 Write 0
2 0 1 - c=0+1=1, 1 < 7: true
3 0 1 0 (0*1)%7
4 0 1 0 Write 0
2 0 2 - c=1+1=2, 2 < 7: true
... ... through c = 6
4 0 6 0 Write 0
2 0 7 - c=+1=7, 7 < 7: false
At this point the inner loop statement, lines 2-4, has completed, and you continue. You go on to the next statement in
the same sequential chuck as the inner loop statement in lines 2-4: That chunk is the the outer loop body, lines 2-6.
The next statement is line 6, advancing printing to the next line. That is the last statement of the outer loop, so you
return to the heading of the outer loop and modify its loop variable r. The two lines just described are:
line r c modProd comment
6 0 - - print a newline
1 1 - - r=s0+1=1, 1 < 7 enter outer loop
Then you go all the way through the inner loop again, for all columns, with c going from 0 through 6, and exit at c=7,
finish the body of the outer loop by advancing to a new print line, and return to the outer loop heading, setting r = 2...,
until all rows are completed.
The common error here is to forget what loop is the innermost one that you are working on, and exit that loop before
is is totally finished. It finishes when the test of the condition controlling the loop becomes false.
Look back one more time and make sure the code for this simpler table makes sense before we continue to the one
with labels....

8.2. Examples With for Statements 111

Introduction to Computer Science in C#, Release 2.0

The fancier table has a couple of extra rows at the top. These two rows are unlike the remaining rows in the body of
the table, so they need special code.
If we go back to our pseudocode we could add to it:
print heading row
print dash-row
for (int r = 0; r < 7; r++) {
print body row
}

First analyze the heading row: Some parts are repetitive and some are not: Print "* |" once, and then there is a
repetitive pattern printing 0 - 6, which we can do with a simpler loop than in the table body:
Console.Write("* | ");
for ( int i = 0; i < 7; i++) {
Console.Write(i + " ");
}
Console.WriteLine();

The dashed line can be generated using StringOfReps from Lab: Loops. How many dashes? For each of seven
columns, and in a row header, we need a digit and an space or (7+1)*(1+1) characters, plus one for the ‘|’: 1 +
(7+1)*(1+1). Thinking ahead, we will leave that expression unsimplified.
We have done most of the work for the rows of the body of the table in the simpler version. We just have a bit
of printing for the initial row label before the column loop. The row label is r. The whole code is in example
mod7_table/mod7_table.cs and below:
Besides the 0 row and 0 column in the mod 7 table, note that in each line the products are a permutation of all the
numbers 1-6. That means it is possible to define the inverse of the multiplication operation, and mod 7 arithmetic
actually forms a mathematical field. A lot more math is useful! Modular arithmetic (with much larger moduli!) is
extremely important in public key cryptography, which protects all your online financial transactions....
The inverse operation to multiplication for prime moduli is easy to work out by brute force, going through the row of
products. A much more efficient method is needed for cryptography: That method involves an elaboration of Greatest
Common Divisor.
Finally, let us generalize this table to mod n. With n up to about 25, it is reasonable to print. Most of the changes are
just replacing 7 by n. There is a further complication with column width, since the numbers can be more than one digit
long. We can do formatting with a field width. Unfortunately in C# the field width must be a literal integer embedded
in the format string, but our number of digits in n is variable.
Here is a good trick: Construct the format string inside the program. We can do that with another format string. To
get the format for a number and an extra space mod 7, we want format string “{0,1} ”, but for mod 11, we want “{0,
2} ”. We can create a format string to substitute into the place where the 1 or 2 goes. This 1 or 2 to substitute is the
number of characters in n as a string, given by ("" + n).Length.
The special format string has an extra wrinkle, because we want explicit braces (for the main format string). Recall
the explicit braces are doubled. Putting this all together, we can create our main format string with:
int numberWidth = ("" + n).Length;
string colFormat = string.Format("{{0,{0}}} ", numberWidth);

The whole function code is below and in example mod_mult_table/mod_mult_table.cs.

Todo
Reversing a string...

Todo

112 Chapter 8. For Loops

 Introduction to Computer Science in C#, Release 2.0

Palindrome

8.2.4 Exercises

Head or Tails Exercise

Write a program heads_tails.cs. It should include a function Flip(), that will randomly prints Heads or
Tails once. Accomplish this by choosing 0 or 1 arbitrarily with a random number generator. Use a static variable
declaration and initialization as in Number Guessing Game Lab:
static Random r = new Random();

Then, for ints low and higher, with low < higher:
int n = r.Next(low, higher);

returns a (pseudo) random int, satisfying low <= n < higher. If you select low and higher as 0 and 2, so
there are only two possible values for n, then you can choose to print Heads or Tails with an if-else statement
based on the result.
In your Main method have a for loop calling Flip() 10 times to test it, so you generate a random sequence of
10 heads and/or tails. With these 10 rapid calls, it is important that a new Random object is only created once. The
suggested static variable declaration ensures that.

Group Flips Exercise

Write a program format_flips.cs. It should include the function Flip(), from the last exercise, and include
another function:
// Print out the results from the total number of random flips of a coin.
// Group them groupSize per line, each followed by a space.
// The last line may contain fewer than groupSize flips
// if total is not a multiple of groupSize. The last line
// should be followed by exactly one newline in all cases.
// For example, GroupFlips(10, 4) *could* produce:
// Heads Heads Tails Heads
// Heads Tails Heads Tails
// Tails Tails
static void GroupFlips(int total, int groupSize)

Test this with a variety of calls to GroupFlips in Main. The previous exercise would correspond to a single call in
Main:
GroupFlips(10, 1);

8.3 Lab: Loops

Goals for this lab:

• Practice with loops. You are encouraged to use a for loop where appropriate.
• Use nested loops where appropriate.
Copy example loop_lab_stub/loop_lab.cs to a new project of yours, and fill in function bodies for each part below:

8.3. Lab: Loops 113

Introduction to Computer Science in C#, Release 2.0

1. Complete
Hint: How would you do something like the example PrintReps("Ok", 9) or with a higher count by
hand? Probably count under your breath as you write:
1 2 3 4 5 6 7 8 9
OkOkOkOkOkOkOkOkOk

This is a counting loop.

2. Complete
Note the distinction from the previous part: Here the function prints nothing. Its work is returned as a single
string. You have to build up the final string.
3. Complete Factorial, in a format much like SumToN in example sum_to_n_test/sum_to_n_test.cs:
It is useful to think of the sequence of steps to calculate a concrete example of a factorial, say 6!:
Start with 1
2 *1 = 2
3*2 = 6
4 * 6 = 24
5*24 = 120
6*120 = 720

ALSO find the largest value of n for which the function works. (You might want to add a bit of code further
testing Factorial, to make this easier.)
4. Modify the function to return a long. Then what is the largest value of n for which the function works?
Remember the values from this part and the last part to tell the TA’s checking out your work.
5. Complete the method
Here are further examples:
PrintRectangle(5, 1, ’ ’, ’B’);
PrintRectangle(0, 2, ’-’, ’+’);

would print
BBBBBBB
B B
BBBBBBB
++
++
++
++

Suggestion: You are always encouraged to build up to a complicated solution incrementally. You might start by
just creating the inner rectangle, without the border.
6. 40% Extra Credit Complete the method below. The comments looks better in the source code, because Sphinx
turns comments to italics, so the vertical bars do not appear vertical below!
Here is further example:
PrintTableBorders(2, 1, 6, 3);

would print (with actual vertical bars)

114 Chapter 8. For Loops

 Introduction to Computer Science in C#, Release 2.0

+------+------+
| | |
| | |
| | |
+------+------+

You can do this with lots of nested loops, or much more simply you can use StringOfReps, possibly six
times, and print a single string. Think of larger and larger building blocks.
The source of this book is plain text where some of the tables are laid out in a format similar to the output of
this function. The Emacs editor has a mode that maintains a fancier related setup on the screen, on the fly, as
content is added inside the cells!

8.3. Lab: Loops 115

Introduction to Computer Science in C#, Release 2.0

116 Chapter 8. For Loops

 CHAPTER

NINE

FILES, PATHS, AND DIRECTORIES

9.1 Files As Streams

Thus far you have been able to save programs, but anything produced during the execution of a program has been lost
when the program ends. Data has not persisted past the end of execution. Just as programs live on in files, you can
generate and read data files in C# that persist after your program has finished running.
As far as C# is concerned, a file is just a string (often very large!) stored on your file system, that you can read or
write, gradually, line by line, or all together.
C# has the abstraction of a stream, as a sequence of characters to be processed sequentially. A stream can either be
written sequentially or read sequentially. You have already read and written streams of characters to the Console. Most
of the syntax that we use for files will be very similar, using methods ReadLine, WriteLine, and Write in the
same way you used them for the Console.
Files can be handled very differently by different operating systems, but C# abstracts away the differences and provides
stream interfaces between a C# program and files.

9.2 Writing Files

Try the following:

1. In Xamarin Studio build, not run, the project first_file. Build is the first selection in the local popup menu for
first_file in the Solution pad. Recall to get the local popup menu * go to the Solution pad * right click on the
project (Mac control-click)
2. Next open an operating system directory window for the project. With Xamarin Studio open, a quick way to do
that is to go to the same popup window, and this time select “Open Containing Folder”.
3. Besides the project files from the Solutions pad, in the directory window you should also see a folder
bin. Change to that folder and then to its sub-folder Debug. This is where the build step put its result
first_file.exe and debug information first_file.exe.mdb. You should see no other file. Leave
this window open.
4. Now, back in Xamarin Studio, run the project. Depending on your operating system, you may or may not see a
Console Window pop up. If you do see one, you should not see any evidence of program results. If you got a
window, close it.
5. Look at the directory window again. You should see a file sample.txt. This is a file created by the program
you ran.
Here is the program:

117
Introduction to Computer Science in C#, Release 2.0

Look at the code. Note the extra namespace being used at the top. You will always need to be using System.IO
when working with files. Here is a slightly different use of a dot, ., to indicate a subsidiary namespace.
The first line of Main creates a StreamWriter object assigned to the variable writer. A StreamWriter links
C# to your computer’s file system for writing, not reading. Files are objects, like a Random, and use the new syntax to
create a new one. The parameter in the constructor gives the name of the file to connect to the program, sample.txt.

Warning: If the file already existed, the old contents are destroyed silently by creating a StreamWriter.

If you do not use any operating system directory separators in the name (’\’ or ’/’, depending on your operating
system), then the file will lie in the current directory, discussed more shortly. The Xamarin Studio default is for this
current directory to be this Debug directory. This will be inconvenient in many circumstances, and later in the chapter
we will see how to minimize the issue.
The second and third lines of Main write the specified strings to lines in the file. Note that the StreamWriter
object writer, not Console, comes before the dot and WriteLine. This is yet another variation on the use of
a dot, .: between an object and a function tied to this object. In this situation the function tied to an object is more
specifically called a method, in object-oriented terminology. All the uses of a dot (except for a numerical literal value)
share a common idea, indicating a named part or attribute of a larger thing.
The last line of Main is important for cleaning up. Until this line, this C# program controls the file, and nothing may
be actually written to the file yet: Since initiating a file operation is thousands of times slower than memory operations,
C# buffers data, saving small amounts and writing a larger chunk all at once.

Warning: The call to the Close method is essential for C# to make sure everything is really written, and to
relinquish control of the file for use by other programs.

It is a common bug to write a program where you have the code to add all the data you want to a file, but the program
does not end up creating a file. Usually this means you forgot to close the file!
Xamarin Studio places sample.txt in a hard-to-guess place in the file system, that is not shown in the Solution
pad, so do not look for it there! You can see it in an operating system file window. The easiest way to do that in
Xamarin Studio is to right-click in the Solution pad on the project, and select Open Containing Folder. This should
start you in the project directory. Then you can drill down through the bin folder to the Debug folder. You can open
the sample.txt file with your favorite text processor. It should contain just what was written!
That folder also contains the executable program created when Xamarin Studio builds your project.
If you were to run the program from the command line instead of from Xamarin Studio, the file would appear in the
current directory.
As you can use a String Format Operation with functions Write and WriteLine, of the Console class, you can
also use a format string with the corresponding methods of a StreamWriter, and embed fields by using braces in
the format string.

9.3 Reading Files

In Xamarin Studio, go to project print_first_file. Right click on the project in the Solution pad, and select Open
Containing Folder. Drill down two folders through bin to Debug. You should find a copy of the sample.txt that
we stored there. You can open it and look at it if you like.
Run the example program print_first_file/print_first_file.cs, shown below:
Now you have read a file and used it in a program.

118 Chapter 9. Files, Paths, and Directories

 Introduction to Computer Science in C#, Release 2.0

In the first line the operating system file (sample.txt) is associated again with a C# variable name (reader), this
time for reading as a StreamReader object. A StreamReader can only open an existing file, so sample.txt,
must already exist.
Again we have parallel names to those used with Console, but in this case the ReadLine method returns the next
line from the file. Here the string from the file line is assigned to the variable line. Each call the ReadLine reads the
next line of the file.
Using the Close method is generally optional with files being read. There is nothing to lose if a program ends without
closing a file that was being read. 1

9.3.1 Reading to End of Stream

In first_file.cs, we explicitly coded reading two lines. You will often want to process each line in a file, without
knowing the total number of lines while programming. This means that files provide us with only our second kind of
a sequence: this time the sequence of lines in the file! To process all of them will require a loop and a new test to
make sure that you have not yet come to the end of the file’s stream: You can use the EndOfStream property. It has
the wrong sense (true at the end of the file), so we negate it, testing for !reader.EndOfStream in the example
program print_file_lines.cs. This little program reads and prints the contents of a file specified by the user,
one line at a time:
var For conciseness (and variety) we declared reader using the more compact syntax with var:
var reader = new StreamReader(userFileName);

You can use var in place of a declared type to shorten your code with a couple of restrictions:
• Use an initializer, from which the type of the variable can be inferred.
• Declare a local variable inside a method body or in a loop heading.
• Declare only a single variable in the statement.
You could have used this syntax long ago, but as the type names become longer, it is more useful!
Things to note about reading from files:
• Reading from a file returns the part read, of course. Never forget the side effect: The location in the file advances
past the part just read. The next read does not return the same thing as last time. It returns the next part of the
file.
• Our while test conditions so far have been in a sense “backward looking”: We have tested a variable that has
already been set. The test with EndOfStream is forward looking: looking at what has not been processed yet.
Other than making sure the file is opened, there is no variable that needs to be set before a while loop testing
for EndOfStream.
• If you use ReadLine at the end of the file, the special value null (no object) is returned. This is not an error,
but if you try to apply any string methods to the null value returned, then you get an error!
Though print_file_lines.cs was a nice simple illustration of a loop reading lines, it was very verbose consid-
ering the final effect of the program, just to print the whole file. You can read the entire remaining contents of a file
as a single (multiline) string, using the StreamReader method ReadToEnd. In place of the reading and printing
loop we could have just had:
string wholeFile = reader.ReadToEnd();
Console.Write(wholeFile);

ReadToEnd does not strip off a newline, like ReadLine does, so we do not want to add an extra newline when
writing. We use the Write method instead of WriteLine.
1 If, for some reason, you want to reread this same file while the same program is running, you need to close it and reopen it.

9.3. Reading Files 119

Introduction to Computer Science in C#, Release 2.0

9.3.2 Example: Sum Numbers in File

We have summed the numbers from 1 to n. In that case we generated the next number i automatically using i++. We
could also read numbers from a file containing one number per line (plus possible white space):
static int CalcSum(string filename)
{
int sum = 0;
var reader = new StreamReader(filename);
while (!reader.EndOfStream) {
string sVal = reader.ReadLine().Trim();
sum += int.Parse(sVal);
}
reader.Close();
return sum;
}

Below and in sum_file/sum_file.cs is a more elaborate, complete example, that also exits gracefully if you give a bad
file name. If you give a good file name, it skips lines that contain only whitespace.
A useful function used in Main for avoiding filename typo errors is in the System.IO namespace is
bool File.Exists(string filenamePath)

It is true if the named files exists in the file system.

You should see the file sum_file/numbers.txt in the Xamarin Studio project. It is in the right form for the program. If
you run the program and enter the response:
numbers.txt

you should be told that the file does not exist. Recall that the executable created by Xamarin Studio is two directories
down through bin to Debug. This is the default current directory when Xamarin Studio runs the program. You can
refer to a file that is not in the current directory. A full description is in the next section, but briefly, what we need now:
The symbol for the parent directory is ... The hierarchy of folders and files are separated by \ in Windows and / on
a Mac, so you can test the program successfully if you use the file name: ..\..\numbers.txt in Windows and
../../numbers.txt on a Mac. On a Mac, running the program looks like:
Enter the name of a file of integers: ../../numbers.txt
The sum is 16

In FIO Helper Class we will discuss a more flexible way of finding files to open, that works well in Xamarin Studio
and many other situations.

Safe Sum File Exercise

1. Copy sum_file.cs to a file safe_sum_file.cs in a new project of yours. Modify the program: Write a
new function with the heading below. Use it in Main, in place of the if statement that checks (only once) for
a legal file:
// Prompt the user to enter a file name to open for reading.
// Repeat until the name of an existing file is given.
// Open and return the file.
public static StreamReader PromptFile(string prompt)

2. A user who forgot the file name woud be stuck! Elaborate the function and program, so that an empty line
entered means “give up”, and null (no object) should be returned. The main program needs to test for this and
quit gracefully in that case.

120 Chapter 9. Files, Paths, and Directories

 Introduction to Computer Science in C#, Release 2.0

9.3.3 Example Copy to Upper Case

Here is a simple fragment from example file copy_upper/copy_upper.cs. It copies a file line by line to a new file in
upper case:
You may test this in the Xamarin Studio example project copy_upper:
1. Expand the copy_upper project in the Solution pad. The project includes the input file. You may not see it at
first. You need to expand the folder for bin and then file:Debug. You see text.txt.
2. To see what else is in the file system directory, you can select the Debug folder and right click and select Open
Containing Folder. You should see text.txt but not upper_text.txt. Leave that operating system file
folder open.
3. Go back to Xamarin Studio and run the project. Now look at the operating system Debug folder again. You
should see upper_text.txt. You can open it and see that it holds an upper case version of the contents of
text.txt.
2
This is another case where the ReadToEnd function could have eliminated the loop.
string contents = reader.ReadToEnd();
writer.Write(contents.ToUpper());

9.4 Path Strings

When a program is running, there is alway a current working directory. When you run a project through Xamarin
Studio, by default the current directory is the directory two levels below the project directory.
Files in the current working directory can to referred to by their simple names, e.g., sample.txt.
Referring to files not in the current directory is more complicated. You should be aware from using the Windows
Explorer or the Finder that files and directories are located in a hierarchy of directories in the file system. On a Mac,
the file system is unified in one hierarchy. On Windows, each drive has its own hierarchy.
Files are generally referred to by a chain of directories before the final name of the file desired. A path string is used
to represent such a sequence of names. Elements of the directory chain are separated by operating system specific
punctuation: In Windows the separator is backslash, \, and on a Mac it is (forward) slash, /. For example on a Mac the
path
/Users/anh

starts with a /, meaning the root or top directory in the hierarchy, and Users is a subdirectory, and anh is a subdirectory
of Users (in this case the home directory for the user with login anh). It is similar with Windows, except there may be
a drive in the beginning, and the separator is a \, so
C:\Windows\System32

is on C: drive; Windows is a subdirectory of the root directory \, and System32 is a subdirectory of Windows. Each
drive in Windows has a separate file hierarchy underneath it.
Paths starting from the root of a file system, with \ or / are called absolute paths. Since there is always a current
directory, it makes sense to allow a path to be relative to the current directory. In that case do not start with the slash
that would indicate the root directory. For example, if the current directory is your home directory, you likely have a
subdirectory Downloads, and the Downloads directory might contain source.zip. From the home directory,
this file could be referred to as Downloads\source.zip or Downloads/source.zip on a Mac.
2 Besides the speed and efficiency of this second approach, there is also a technical improvement: There may or may not be a newline at the

end of the very last line of the file. The ReadLine method works either way, but does not let you know the difference. In the line-by-line version,
there is always a newline after the final line written with WriteLine. The ReadToEnd version will have newlines exactly matching the input.

9.4. Path Strings 121

Introduction to Computer Science in C#, Release 2.0

Relative to a Xamarin Studio project directory, the current directory for execution of the program is bin\Debug or
bin/Debug on a Mac.
Referring to files in the current directory just by their plain file name is actually an example of using relative paths.
With relative paths, you sometimes want to move up the directory hierarchy: .. (two periods) refers to the directory
one level up the chain.
Next imagine reversing the relative path from a Xamarin Studio project directory to the current directory for execution:
If the current directory is the execution directory, then .. refers to directory bin, and then ..\.. or ../.. refers
to the project directory. Further, if the project directory contains the file numbers.txt, then it could be referred to
relative to the execution directory as ..\..\numbers.txt or ../../numbers.txt.
Occasionally you need to refer explicitly to the current directory: It is referred to as ”.” (a single period).

9.4.1 Paths in C#

The differing versions of paths for Windows and a Mac are a pain to deal with. Luckily C# abstracts away the
differences. It has a Path class in the System.IO namespace that provides many handy functions for dealing with
paths in an operating system independent way:
For one thing, C# knows the path separator character for your operating system,
Path.DirectorySeparatorChar.
More useful is the function Path.Combine, which takes any number of string parameters for sequen-
tial parts of a path, and creates a single string appropriate for the current operating system. For ex-
ample, Path.Combine("bin", "Debug") will return "bin\Debug" or "bin/debug" as appro-
priate. Path.Combine("..", "..", "numbers.txt") will return "..\..\numbers.txt" or
"../../numbers.txt".
You can look at the Path class in the MSDN documentation for many other operations with path strings.
Path strings are used by the Directory Class and by the File Class.

9.5 Directory Class

The Directory class is in the System.IO namespace. Directories in the file system are referenced by Path Strings.
You can look at the MSDN documentation for a wide variety of functions in the Directory class including ones to
list all the files in a directory or to check if a path string represents an actual directory.

9.6 File Class

We will generally access operating system files using the Stream abstraction, discussed earlier in this chapter. There
is also a File class is in the System.IO namespace, with a number of specialized and convenience functions. We
already used the File.Exists function.
You can look at the MSDN documentation for other uses of the File class.

9.7 Command Line Execution

C# shields you from the differences between operating systems with its File, Path, and Directory classes.

122 Chapter 9. Files, Paths, and Directories

 Introduction to Computer Science in C#, Release 2.0

If you leave Xamarin Studio and go to the command line as described in Command Line Introduction, then you are
exposed to the differences between the operating systems. Look over that section.
Thus far we have let Xamarin Studio hide what actually is happening when you execute a program. The natural
environment for the text-based programs we are writing is the command line. We need to get outside Xamarin Studio.
1. To show off the transition, first build or run the addition1 example project from inside Xamarin Studio.
2. Open a terminal on a Mac or a Mono Command Prompt console in Windows.
3. Following the Command Line Introduction, change the current directory to the addition1 example project direc-
tory.
4. Then change to the subdirectory bin and then the subdirectory Debug.
5. Enter the command to list the directory (dir in Windows; ls on a Mac).
6. You should see addition1.exe. This is the compiled program created by Xamarin Studio. Enter the com-
mand
mono addition1.exe

This should run your program. Note that when you complete it, the window does not disappear! You keep that
history. Keep this terminal/console window open until the end of the chapter.
7. Windows only: On Windows, Xamarin Studio creates a regular Windows executable file. For consistency you
can use the command above, but you no longer need Mono. You can just enter the command addition1.exe
or the shorter addition1.

9.7.1 MCS: Compiling

Continue with the same terminal/console window. Let us now consider creating an executable program for
addition1.cs, directly, without using Xamarin Studio:
1. Enter the command cd .. and then repeat: cd .., to go up to the project directory.
2. Print a listing of the directory. You should see addition1.cs but not addition1.exe.
3. Try the command
mono addition1.exe

You should get an error message, because addition1.exe is not in the current directory.
4. Enter the command
mcs addition1.cs

This is the Mono system compiler, building from the source code.
5. Print a listing of the directory. You should see now addition1.exe, created by the compiler.
6. Try the command again:
mono addition1.exe

Now try a program that had multiple files. The project version addition3 uses the library class UIF. Continue with the
same terminal/console window:
1. Enter the commands:
cd ../addition3
mcs addition3.cs

9.7. Command Line Execution 123

Introduction to Computer Science in C#, Release 2.0

You should get an error about missing the UIF class. The mcs program does not know about the information
Xamarin Studio keeps in its references.
2. Extend the command:
mcs addition3.cs ../ui/uif.cs

That should work, now referring to both needed files.

3. Enter the command
mono addition3.exe

4. Now let us try a project where we read a file. Enter commands

cd ../sum_file
mcs sum_file.cs
mono sum_file.exe

We ran this program earlier through Xamarin Studio. Recall that that entering the file name numbers.txt
failed, and to refer to the right place for the numbers.txt file, we needed to use ..\..\numbers.txt or
../../numbers.txt. This time just enter numbers.txt. The program should work, giving the answer
16.
By default mcs and mono read from and write to the current directory of the terminal/console. In the situation above,
sum_file.cs and numbers.txt were in the project directory, which is the current directory. Then sum_file.exe
was written to and run from the same directory.
This is unlike the Xamarin Studio default, where the current directory for execution is not the project directory.
Under the hood, Xamarin Studio uses mcs also, with a bunch of further options in the parameters, changing the
execution directory and also arranging for better debugging information when you get a runtime error.
Xamarin Studio keeps track of all of the parts of your projects, and recompiles only as needed. There are also
command-line tools that manage multi-file projects neatly, remembering the parts, and compiling only as necessary.
One example is NAnt, which comes with Mono.

9.8 FIO Helper Class

We have already discussed and used the UI class to aid keyboard input. Now we are going to develop an FIO class for
our libraries. The FIO class aids file input and output with Xamarin Studio, and illustrates a number of more generally
useful ideas.
You saw in the last section how we might refer to numbers.txt in different ways depending on the execution
environment. Our situation is based on the particular choices made by the creators of Xamarin Studio. More generally,
there are many times when a program may need a file that may be stored in one of several directories.
Our FIO class will address this issue, and we will set up the parameters to work specifically with both Xamarin Studio
and command line development.
We use one idea that is discussed more in the next chapter: We need a sequence of directory strings to look through.
At this point we have only discussed sequences of individual characters. The variable paths contains a sequence
of directory paths to check. In our case we make the sequence contain ".", the current directory, "..", the parent
directory, and Path.Combine("..", ".."), the parent’s parent. We make paths a static variable, so it is
visible in all the functions in the class.
Then the sequence paths can be used in the foreach loop:

124 Chapter 9. Files, Paths, and Directories

 Introduction to Computer Science in C#, Release 2.0

For each directory path in paths, we create a filePath as if the file were in that directory. We return the first path
that actually exists. We allow for the file to not be in any of the directories in paths. If we do not find it, we return
null (no object).
For convenience, we have an elaboration, using GetPath, that directly opens the file to read:
We have a variation on GetPath that just return the path to the directory containing the file. Here is the heading:
This is useful in case you want to later write into the same directory that you read from. You can get a location from
GetLocation and then write to the same directory, creating a StreamWriter. Use the convenience function:
The entire FIO class is in fio/fio.cs
We illustrate the use of FIO functions in example file fio_usage/fio_usage.cs:
IF you look at the fio_usage project in our examples solution, you see that sample.txt is a file in the project folder.
The program ends up writing to a new file in the same (project) directory. Remember that even though the new file
output.txt appears in the project directory, it does not appear in the Solution pad unless you add it to the project.
You can see it in the file system, and open it if you like.
If you want to open a terminal/console and go to the project directory, you can compile and run this program, and it
will still work, even though the current directory has changed.
You are encouraged to make a library project fio in your work solution, copying the fio.cs file. (Follow instruction like
for ui in Library Projects in Xamarin Studio.) You can test your new library by also copying the fio_test project to
your solution. If you do this now and stick to one work solution, then you will be ready for several later uses of FIO.

9.8.1 File Line Removal Exercise

Complete the function described below, and make a Main program and sample file to test it:
/// Take all lines from reader that do not start with startToRemove
/// and copy them to writer.
static void FileLineRemoval(StreamReader reader, StreamWriter writer
char startToRemove)

For example, in Unix/Mac scripts lines starting with ’#’ are comment lines. Making startToRemove be ’#’
would write only non-comment lines to the writer.

9.8. FIO Helper Class 125

Introduction to Computer Science in C#, Release 2.0

126 Chapter 9. Files, Paths, and Directories

 CHAPTER

TEN

ARRAYS

10.1 One Dimensional Arrays

10.1.1 Basic Syntax

A string is an immutable sequence of characters. Arrays provide more general sequences, with the same indexing
notation, but with free choice of the type of the items in the sequence, and the ability to change the elements in the
sequence.
For example, if we want the type for an array with int elements, it is int[]. In general for any element type, the
type for an array of the element type is
type[]
so
int[] a;

declares a to refer to an array containing int elements. You do not know how many elements will be allowed in this
array from this declaration. We must give further information to create the corresponding array object. All object can
be created using the new syntax. An array must get a definite length, which can be a literal integer of any integer
expression. For example
int[] a;
a = new int[4];

or combined with the declaration,

int[] a = new int[4];

creates an array that holds 4 integers. The elements of the array must get initial values. Numerical arrays get initialized
to all 0’s with this syntax.
For a variety of reasons, including bookkeeping by the compiler, the actual data for an array is not stored directly in
the memory location allocated by the declaration. The array could have any number of items, and hence the memory
requirements are not known at compile time. Like all other object (as opposed to primitive) types, what is actually
stored at the memory location declared for a is a reference to the actual place where the data for the array is stored. In
actual compiler implementation this reference is an address in memory. In diagrams we will illustrate object references
with an arrow pointing to the actual location for the object’s data. For example after a is initialized:

127
Introduction to Computer Science in C#, Release 2.0

The small box beside a is meant to indicate the memory space allocated when a is declared. As you can see that space
does not actually contain the array, but only a reference to the array, pointing to the actual sequence of data for the
array. To make it easy to refer to the elements in the diagram, we also label the indices associated with each element,
though they are not actual a part of what is stored in memory.
The general syntax to create a new array is
new type[ length ]
After the type, there are square brackets enclosing an expression for the length of the array - this length is unchangeable
after creation.
The elements inside an array can to referenced with the same index notation used earlier for strings.
a[2]

refers to the element at index 2 (third element because of 0 based indexing).

Unlike with strings, this element can not only be read, but also be assigned to:
a[0] = 7;
a[1] = 5;
a[2] = 9;
a[3] = 6;

These four assignment statements would replace the original 0 values for each element in the array.
This is a verbose way to specify all array values. An array with the same final data could be created with the single
declaration:
int[] b = {7, 5, 9, 6};

The list in braces ONLY is allowed as an initialization of a variable in a declaration, not in a later assignment statement.
Technically it is an initializer, not an array literal.
Individual array elements can both be used in expressions, and be assigned to. Continuing with the earlier example
code:
a[2] = 4*a[1] - a[3];

a[2] now equals 4*5 - 6 = 14.

Arrays, like strings, have a Length property:
Console.WriteLine(b.Length); // prints 4

Just as we saw that using a variable for an index was useful with strings, array elements are almost always referred to
with an index variable in practice. A very common pattern is to deal with each element in sequence, and the syntax is
the same as for a string. Print all elements of array b:
for (int i= 0; i < b.Length, i++) {
Console.WriteLine(b[i]);
}

You could also use while syntax. The foreach syntax would be:

128 Chapter 10. Arrays

 Introduction to Computer Science in C#, Release 2.0

foreach( int x in b) {
Console.WriteLine(x);
}

The int type for x matches the element type of the array b.
The shorter foreach syntax is not as general as the for syntax. For example, to print only the first 3 elements of b:
for(int i= 0; i < 3; i++) {
Console.WriteLine(b[i]);
}

but the foreach syntax would not work, since it must process all elements.
Also use the for syntax to assign new values to the array elements, rather than just use the values in expressions:
for(int i= 0; i < b.Length; i++) {
b[i] = 5*i;
}

Now the array b of our earlier examples (of length 4) would contain 0, 5, 10, and 15.

10.1.2 Parameters to Main

The Main function may take an array of strings as parameter, as in example print_param/print_param.cs:
By convention, the formal parameter for Main is called args, short for arguments.
Compile and run the program from the command line. Run it again with some things at the end of the line like:
mono print_param.exe hi there 123

This should print for you:

There are 3 command line parameters.
hi
there
123

See what quoted strings do. Use command line parameters (with the quotes) "hi there" 123. This should print
for you:
There are 2 command line parameters.
hi there
123

You can simulate command line parameters inside Xamarin Studio:

1. Open the local popup menu for the project you are using.
2. Select Run With >
3. In the submenu select Custom Parameters.
4. That brings up a dialog where you can enter the desired command line parameters.
5. Optionally you can remember this setup by clicking on box in front of “Save this configuration as a custom
execution mode”. If you check it, you get a place to enter a Custom Mode Name.
6. End up clicking the Execute button.
7. If you set a Custom Mode, later when you get to the submenu after “Run With >”, you will see your custom
mode name to select!

10.1. One Dimensional Arrays 129

Introduction to Computer Science in C#, Release 2.0

Try it!

10.1.3 String Method Split

A string method producing an array:

string[] Split(char separator ) Returns an array of substrings from this string. They are the pieces left after
chopping out the separator character from the string. A piece may be the empty string. Example:
csharp> var fruitString = "apple pear banana";
csharp> string[] fruit = fruitString.Split(’ ’);
csharp> fruit;
{ "apple", "pear", "banana" }
csharp> fruit[1];
"pear"
csharp> var s = " extra spaces ";
csharp> s.Split(’ ’);
{ "", "", "extra", "", "", "spaces", "" }

Note: The response with the list in braces is a purely csharp convention for displaying sequences for the user. There is
no corresponding string displayed by C# Write commands. Also see that the string is split at each separator, even
if that produces empty strings.
Split is useful for parsing a line with several parts. You might get a group of integers on a line of text, for instance
from:
string input = UI.PromptLine(
"Please enter some integers, separated by single spaces: ");

To extract the numbers, you want to the separate the entries in the string with Split, and you probably want further
processing: If you want them as integers, not strings, you must convert each one separately.
It is useful to put this idea in a function. See the type returned. It is an array int[] for the int results:
In a call to IntsFromString("2 5 22"), integers would be an array containing strings “2”, “5”, and “22”.
We need the conversions to int to go in a new array that we call data. We must set its length, which will clearly be
the same as for integers, integers.Length. To assign elements into data we need a loop providing indices,
like the for loop provided. Then for each index, we parse a string in integers into an int, and place the int in
the corresponding location in data. We need to return data at the end to make it accessible to the caller.
Remember some patterns illustrated here, that you will use over and over:
• You can return any type, including an array type.
• If you want to return a new array, you must first create an array of the proper length before you can make
assignments to individual elements.
• The use of the same index variable in more than one array is a standard way to have related entries in corre-
sponding positions of the arrays.
We will use this function for testing in Linear Searching.

GetTokens Exercise

Complete this function:

///Return an array with the string split at blanks
/// into positive length tokens.
/// Example: GetTokens(" extra spaces ") returns an

130 Chapter 10. Arrays

 Introduction to Computer Science in C#, Release 2.0

/// array containing {"extra", "spaces"}

public static string[] GetTokens(string s)

Hint: It is harder now to get the right length for your new array: After splitting at ’ ’, you can count the non-empty
strings in the result. Then create your new array and copy only non-empty strings. Handling the indices for the new
array also gets more complicated.

10.1.4 References and Aliases

Object variables, like arrays, are references, and this has important implications for assignment.
With a primitive type like an int, an assignment copies the data:

In the diagram, the contents of the memory box labeled a is copied to the memory box labeled d. The value of d starts
off equal to the value of a, but can later be changed independently.
Contrast an assignment with arrays. The value that is copied is the reference, not the array data itself, so both end up
pointing at the same actual array:

Hereafter, array assignments like:

b[2] = -10;
d[1] = 55;

would both change the same array. Now b and d are essentially names for the same thing (the actual array). The
technical term matches English: The names are aliases.
This may seem like a pretty silly discussion. Why bother to give two different names to the same object? Isn’t one
enough? In fact it is very important in function/method calls. An array reference can be passed as an actual value, and
it is the array reference that is copied to the formal parameter, so the formal parameter name is an alias for the actual
parameter name.

Note: If an array passed as a parameter to a method has elements changed in the method, then the change affects the
actual parameter array. The change remains in the actual parameter array after the method has terminated.

For example, consider the following function:

// Modify a by multiplying all elements by multiplier.
static void Scale(int[] a, int multiplier)
{

10.1. One Dimensional Arrays 131

Introduction to Computer Science in C#, Release 2.0

for (int i = 0; i < a.Length; i++) {

a[i] *= multiplier; // or: a[i] = a[i] * multiplier
}
}

The fragment:
int[] nums = {2, 4, 1};
Scale(nums, 5);

would change nums, so it ends up containing elements 10, 20, and 5.

10.1.5 Anonymous Array Initialization

Sometimes you only want to use an array with specific values as a parameter to a function. You could write something
like
int[] temp = {3, 1, 7};
SomeFunc(temp);

but if temp is never going to be referenced again, you can do this without using a name:
SomeFunc(new int[] {3, 1, 7});

Like with the use of var, the compiler can infer the type of the array, and the last example could be shortened to
SomeFunc(new[] {3, 1, 7});

It is essential to include the new int[] or new[], not just the {3, 1, 7}.
Such an approach could also be used if you want to return a fixed length array, where you have values for each parts,
as in:
int minVal = ...
int maxVal = ...

return new[] {minVal, maxVal};

10.1.6 Default Initializations

You should have notice that when the first example array of integers was created, it was filled with zeros. It is a safety
feature of C# that the internal fields of objects always get a specific value, not random data. Here are the defaults:
primitive numeric types: zero
boolean: false
all object types: null

132 Chapter 10. Arrays

 Introduction to Computer Science in C#, Release 2.0

Warning: An array with elements of object type, like string[], without a specific initializer, gets initialized to
all null values. The creation is totally legal, but if you try to use the created value, like

string[] words = new string[10];

Console.WriteLine(words[0].Length); // run time error here

The error is because null is not an object - it does not have a Length property. If, for example, you want an
array of empty strings you would need a loop:

for (int i = 0; i < words.length; i++) {

words[i] = "";
}

Command Line Adder Exercise

Write a program adder.cs that calculates and prints the sum of command line parameters, so if you make the
command line parameters in Xamarin Studio be
2 5 22

then the program prints 29.

Do try running from the command line: If you compiled with Xamarin Studio, that means going down to the bin/Debug
directory. Recall Xamarin Studio for Windows produces a Windows executable, not a Mono file, so you can run
adder 2 5 22

while on a Mac you need to run with mono:

mono adder.exe 2 5 22

Trim All Exercise

Write a program trimmer.cs that includes and tests a function with heading:
// Trim all elements of s and replace them in the array.
// Example: If a contains {" is ", " it", "trimmed? "}
// then after the function call the array contains
// {"is", "it", "trimmed?"}.
static void TrimAll(string[] a)

Count Duplicates Exercise

Write a program count_dups.cs that includes and tests a function with heading:
// Return the number of duplicate pairs in an array a.
// Example: for elements 2, 5, 1, 5, 2, 5
// the return value would be 4 (one pair of 2’s three pairs of 5’s.
public static int dups(int[] a)

Mirror Array Exercise

Write a program make_mirror.cs that includes and tests a function with heading:

10.1. One Dimensional Arrays 133

Introduction to Computer Science in C#, Release 2.0

// Create a new array with the elements of a in the opposite order.

// {"aA", "bB", "cC"} produces a new array {"cC", "bB", "aA"}
public static string[] Mirror(string[] a)

Reverse Array Exercise

Write a program reverse_array.cs that includes and tests a function with heading:
// Reverse the order of array elements.
// {"aA", "bB", "cC"} -> {"cC", "bB", "aA"}
public static void Reverse(string[] a)

Histogram Exercise

Write a program make_histogram.cs that includes and tests a function with heading:
// Return a histogram array counting repetitions of values
// start through end in array a. The count for value start+i
// is in index i of the returned array. For example:
// Histogram(new int[]{2, 0, 3, 5, 3, 5}, 0, 5) returns
// a new array containing {1, 0, 1, 2, 0, 2}.
public static int[] Histogram(int[] a, int start, int end)

10.2 Musical Scales and Arrays

Music in the western classical tradition uses a twelve-tone chromatic scale. Any of the tones in this scale can be the
basis of a major scale. Most musicians (especially pianists) learn the C-major scale in the early days of study, owing
to the ability to play this scale entirely with the ivory (white) keys.
The following declaration shows how to initialize an array consisting of the twelve tones of the chromatic scale,
starting from the C note.
Even if you’re not a musician, learning the basic principles is fairly straightforward.
The well-known C-major scale, which is often sung as:
Do Re Mi Fa So La Ti Do

has the following progression:

C D E F G A B C

This progression is known as the diatonic major scale. If you look at the tones array, you can actually figure out the
intervals associated with this array:
C + 2 = D
D + 2 = E
E + 1 = F
F + 2 = G
G + 2 = A
A + 2 = B
B + 1 = C

So given any starting note, the major scale can be generated from the intervals (represented as an array).

134 Chapter 10. Arrays

 Introduction to Computer Science in C#, Release 2.0

So, for example, if you want the F-major scale, you can get it by starting at F and applying the steps of 2, 2, 1, 2, 2, 2,
1:
F + 2 = G
G + 2 = A
A + 1 = B’ (flat) a.k.a. A#)
B’+ 2 = C
C + 2 = D
D + 2 = E
E + 1 = F

So this is the F-major scale:

F G A B’ C D E F

We begin by creating a helper function, FindTone(), which does a linear search to find the key of the scale we want
to compute. The aim is to make it easy for the user to just specify the key of interest. Then we can use this position to
compute the scale given the major (or minor, covered shortly) interval array.
To see what this function does, pick your favorite key (C and G are very common for beginners).
• FindTone("C") gives 0, the first position in the tones array.
• FindTone("G") gives 8.
For example, C is the first note in the array of tones, so FindTone("C") would give us 0. FindTone("F") would
give us 6.
So let’s take a look at ComputeScale() which does the work of computing a scale, given a key and an array of
steps. The scale array is allocated by the Main() method, primarily to allow the same array to be used repeatedly for
calculating other scales.
1. The first thing to note is the setup of this code. We’re going to keep the startTone (obtained by calling
FindTone()) and tonePosition, which is the note we are presently visiting in the tones array.
2. Remember that every scale (e.g. C, D, F#, etc.) can always be obtained by looking at tones and using the
appropriate intervals (the steps parameter) to compute the next note, given a current note.
3. We do some simple checks in line 6 (to ensure that a valid key was specified by the caller) and in line 8 to ensure
that the number of steps + 1 is the length of the scale–and the length of the scale is 8. (We technically don’t have
to limit the scale to 8, because scales can keep going until you run out of playable notes on the instrument.)
4. We’ll now start at the initial position (where we found the base note of the key) and enter a for loop to compute
all of the notes in the scale. This loop iterates over the entries in the steps array to decide what the next note
is.
5. The next note in the scale, scale[i] is computed by taking tonePosition % tones.GetLength(0).
We need to do this, because in most scales, you will eventually end up “falling off the end” of the tones array,
which mens that you need to continue computing notes from the beginning of the array. You can inspect this for
yourself by picking a scale (say, B) that is starting at the end of the tones array. This means you will need to
go to the beginning of the array to get C# (which is 2 tones away from B).
6. The next note is found by adding steps[i] to tonePosition.
The following function writes the scale out (rather naively) by just printing the notes from our existing tones array.
We say that the output is naive because any musician will tell you that a scale should be printed in a normalized way.
For example, the F-major scale (shown above in our earlier explanation) is never written with A# as one of its notes.
It is written as B-flat. It’s easy to manage the various cases by consulting the circle of fifths, which gives us guidance
on the number of flats/sharps each scale has.
Lastly, we put this all together.

10.2. Musical Scales and Arrays 135

Introduction to Computer Science in C#, Release 2.0

This Main() method shows how to set up the steps for both major and minor scales. We’ve already explained how
to express the steps of a major scale. The minor scale basically drops the 3rd and 7th by a semitone (a single step),
which gives us a different pattern.
You can run this program to see the major and minor scales.

10.3 Linear Searching

In this section, we’ll take a look at how to search for a value in an array. Although a fairly straightforward topic, it is
one that comes up repeatedly in programming.
These examples make use of arrays and loops, not to mention functions (for putting it all together).

10.3.1 Linear Search

By far, one of the most common searches you will see in typical programs. It also happens to be one of the more
misused searches, which is another reason we want you to know about it.
Here is the code from example searching/searching.cs to perform a linear search for an integer in an array:
Here’s what it does:
• In lines 5-6 we set up a loop to go from 0 to N-1. We often use N to indicate the size of the array (and it’s much
easier to type than data.Length.
• In line 7, we see whether we found a match for the item we are searching. If we find the match, we immediately
leave the loop by returning the position where it was found.
• It is worth noting here that the array, data, may or my not be in sorted order. So our search reports the first
location where we found the value. It is entirely possible that the more than one position in the array contains
the matching value. If you wanted to find the next one, you could modify the IntArrayLinearSearch()
method to have a third parameter, start, that allows us to continue searching from where we left off. It might
look something like the following:
The following code shows how to use the linear search:
In this example, we ask the user to enter an array of data by entering the values space separated on a line. . To convert
to an int array we use the function IntsFromString discussed in String Method Split.
To allow easy termination of the testing loop, we do not use PromptInt for searchItem, because any int could
be the search target. By using PromptLine, we can allow an empty string as the response, and we test for that to
terminate the loop.
The rest is mostly self-explanatory.

10.4 Sorting Algorithms

Sorting algorithms represent foundational knowledge that every computer scientist and IT professional should at least
know at a basic level. And it turns out to be a great way of learning about why arrays are important.
In this section, we’re going to take a look at a number of well-known sorting algorithms with the hope of sensitizing
you to the notion of performance–a topic that is covered in greater detail in courses such as algorithms and data
structures.
This is not intended to be a comprehensive reference at all. The idea is to learn how these classic algorithms are
coded in the teaching language for this course, C#, and to understand the essentials of analyzing their performance,

136 Chapter 10. Arrays

 Introduction to Computer Science in C#, Release 2.0

both theoretically and experimentally. For a full theoretical treatment, we recommend the outstanding textbook by
Niklaus Wirth [WirthADP], who invented the Pascal language. (We have also adapted some examples from Thomas
W. Christopher’s [TCSortingJava] animated sorting algorithms page.

10.4.1 Exchanging Array Elements

We’ll begin by introducing you to a simple method, whose only purpose in life is to swap two data values at positions
m and n in a given integer array:
For example if we have an array nums, shown with indices:
nums: -1 8 11 22 9 -5 2
index: 0 1 2 3 4 5 6

Then after Exchange(nums, 2, 5) the array would look like

nums: -1 8 -5 22 9 11 2
index: 0 1 2 3 4 5 6

In general, swapping two values in an array is no different than swapping any two integers. Suppose we have the
following integers a and b:
int a, b;
int t;

a = 25;
b = 35;
t = a;
a = b;
b = t;

After this code does its job, the value of a would be 35 and the value of b would be 25.
So in the Exchange() function above, if we have two different array elements at positions m and n, we are basically
getting each value at these positions, e.g. data[m] and data[n] and treating them as if they were a and b in the
above code.
You might find it helpful at this time to verify that the above code does what we’re saying it does, and a good way is
to type it directly into the C# interpreter (csharp) so you can see it for yourself.
The Exchange() function is vital to all of the sorting algorithms in the following way. It is used whenever two items
are found to be out of order. When this occurs, they will be swapped. This doesn’t mean that the item comes to its
final resting place in the array. It just means that for the moment, the items have been reordered so we’ll get closer to
having a sorted array.
Let’s now take a look at the various sorting algorithms.

10.4.2 Bubble Sort

The Bubble Sort algorithm works by repeatedly scanning through the array exchanging adjacent elements that are out
of order. Watching this work with a strategically-placed Console.WriteLine() in the outer loop, you will see
that the sorted array grows right to left. Each sweep picks up the largest remaining element and moves to the right as
far as it can go. It is therefore not necessary to scan through the entire array each sweep, but only to the beginning of
the sorted portion.
We define the number of inversions as the number of element pairs that are out of order. They needn’t be adjacent.
If data[7] > data[16], that’s an inversion. Every time an inversion is required, we also say that there is corre-

10.4. Sorting Algorithms 137

Introduction to Computer Science in C#, Release 2.0

sponding data movement. If you look at the Exchange() code, you’ll observe that a swap requires three movements
to take place, which happens very quickly on most processors but still amounts to a significant cost.
There can be at most 𝑁 · 𝑁 2−1 inversions in the array of length 𝑁 . The maximum number of inversions occurs when
the array is sorted in reverse order and has no equal elements.
Bubble Sort exchanges only adjacent elements. Each such exchange removes precisely one inversion; therefore,
Bubble Sort requires 𝑂(𝑁 2 ) exchanges.

10.4.3 Selection Sort

The Selection Sort algorithm works to minimize the amount of data movement, hence the number of Exchange()
calls.
It’s a remarkably simple algorithm to explain. As shown in the code, the actual sorting is done by a function,
IntArraySelectionSort(), which takes an array of data as its only parameter, like Bubble sort. The way
Selection Sort works is as follows:
1. An outer loop visits each item in the array to find out whether it is the minimum of all the elements after it. If it
is not the minimum, it is going to be swapped with whatever item in the rest of the array is the minimum.
2. We use a helper function, IntArrayMin() to find the position of the minimum value in the rest of the array.
This function has a parameter, start to indicate where we wish to begin the search. So as you can see from the
loop in IntArraySelectionSort(), when we start with position i, and we compare to the later elements
from position i + 1 to the end of the array, updating the position of the smallest element so far.
An illustration to accompany the discussion:
data: 12 8 -5 22 9 2
index: 0 1 2 3 4 5
i k

The first time through the loop, i is 0, and k gets the value 2, since data[2] is -5, the smallest element. Those two
positions get swapped.
data: -5 8 12 22 9 2
index: 0 1 2 3 4 5

The next time through the loop i is 1 and k becomes 5:

data: -5 8 12 22 9 2
index: 0 1 2 3 4 5
i k

After the swap:

data: -5 2 12 22 9 8
index: 0 1 2 3 4 5

and so on. Here is the data after each of the last three swaps:
-5 2 8 22 9 12

-5 2 8 9 22 12

-5 2 8 9 12 22

Consider the first call to IntArrayMin.

data: 12 8 -5 22 9 2
index: 0 1 2 3 4 5

138 Chapter 10. Arrays

 Introduction to Computer Science in C#, Release 2.0

Initially minPos is 0. Here are the changes for each value of pos:
pos=1: 8 = data[1] < data[0] = 12, so minPos becomes 1
pos=2: -5 = data[2] < data[1] = 8, so minPos becomes 2
pos=3: 22 = data[3] is not < data[2] = -5, so minPos still 2
pos=4: 9 = data[4] is not < data[2] = -5, so minPos still 2
pos=5: 2 = data[5] is not < data[2] = -5, so minPos still 2

and 2 gets returned, and we ge the swap

data: -5 8 12 22 9 2
index: 0 1 2 3 4 5

The next call to IntArrayMin has start as 1, so minPos is initially 1, and we compare to the elements of data
at index 2, 3, 4, and 5....
We won’t do the full algorithmic analysis here. Selection Sort is interesting because it does most of its work through
comparisons, with the same number of them no matter how the data are ordered, exactly 𝑁 · 𝑁 2−1 , which is 𝑂(𝑁 2 ) The
number of exchanges is O(N). The comparisons are a non-trivial cost, however, and do show in our own performance
experiments with randomly-generated data.

10.4.4 Insertion Sort

In the Insertion Sort algorithm, we build up a longer and longer sorted list from the bottom of the array. We repeatedly
insert the next element into the sorted part of the array by sliding it down (using our familiar Exchange() method)
to its proper position.
Consider the earlier example array as we illustrate some of the steps. I use the symbol ‘-‘ for an element that we know
to be in the sorted list at the beginning of the array, and ‘@’ over the next one we are trying to insert at the right
position. We start with a one-element sorted list and try to position the second element:
- @
data: 12 8 -5 22 9 2
index: 0 1 2 3 4 5

After each outer loop in sequence we end up with:

- - @
data: 8 12 -5 22 9 2
index: 0 1 2 3 4 5

- - - @
data: -5 8 12 22 9 2
index: 0 1 2 3 4 5

- - - - @
data: -5 8 12 22 9 2
index: 0 1 2 3 4 5

- - - - - @
data: -5 8 9 12 22 2
index: 0 1 2 3 4 5

- - - - - -
data: -5 2 8 9 12 22
index: 0 1 2 3 4 5

Let us illustrate several times just through the inner loop, the first time, when the 8 is moved into position from index
j = 1, so i starts at 1. We show the letter i over the data at index i, and show the comparison test to be done with a >

10.4. Sorting Algorithms 139

Introduction to Computer Science in C#, Release 2.0

with a ‘?’ over it.

? i
data: 12 > 8 -5 22 9 2 1>0 and 12>8: true, so swap, loop
index: 0 1 2 3 4 5

i
data: 8 12 -5 22 9 2 0>0 false, skip comparison of data, end loop
index: 0 1 2 3 4 5

Let us also illustrate at a later time through the inner loop, when the 9 is moved into position from index j = 4, so i
starts at 4.
? i
data: -5 8 12 22 > 9 2 4>0 and 22>9: true, so swap, loop
index: 0 1 2 3 4 5

? i
data: -5 8 12 > 9 22 2 3>0 and 12>9: true, so swap, loop
index: 0 1 2 3 4 5

? i
data: -5 8 > 9 12 22 2 2>0 and 8>9: false, end inner loop
index: 0 1 2 3 4 5

The 9 started at index j = 4, and now the list is sorted up through index 4.
This will require as many exchanges as Bubble Sort, since only one inversion is removed per exchange. So Insertion
Sort also requires 𝑂(𝑁 2 ) exchanges. On average Insertion Sort requires only half as many comparisons as Bubble
Sort, since the average distance an element must move for random input is one-half the length of the sorted portion.

10.4.5 Shell Sort

Shell Sort is basically a trick to make Insertion Sort run faster. If you take a quick glance at the code and look beyond
the presence of two additional outer loops, you’ll notice that the code looks very similar.
Since Insertion Sort removes one inversion per exchange, it cannot run faster than the number of inversions in the
data, which in worst case is 𝑂(𝑁 2 ). Of course, it can’t run faster than N, either, because it must look at each element,
whether or not the element is out of position. We can’t do any thing about the lower bound O(N), but we can do
something about the number of steps to remove inversions.
The trick in Shell Sort is to start off swapping elements that are further apart. While this may remove only one
inversion sometimes, often many more inversions are removed with intervening elements. Shell Sort considers the
subsequences of elements spaced k elements apart. There are k such sequences starting at positions 0 through k-1 in
the array. In these sorts, elements k positions apart are exchanged, removing between 1 and 2(k-1)+1 inversions.
Swapping elements far apart is not sufficient, generally, so a Shell Sort will do several passes with decreasing values
of k, ending with k=1. The following examples experiment with different series of values of k.
In this first example, we sort all subsequences of elements 8 apart, then 4, 2, and 1. Please note that these intervals are
to show how the method works–not how the method works best.
In general, shell sort with sequences of jump sizes that are powers of one another doesn’t do as well as one where
most jump sizes are not multiples of others, mixing up the data more. In addition, the number of intervals must be
increased as the size of the array to be sorted increases, which explains why we allow an arbitrary array of intervals
to be specified.
Without too much explanation, we show how you can choose the intervals differently in an improved shell sort, where
the intervals have been chosen so as not to be multiples of one another.

140 Chapter 10. Arrays

 Introduction to Computer Science in C#, Release 2.0

Donald Knuth has suggested a couple of methods for computing the intervals:

ℎ0 = 1
ℎ𝑘+1 = 3ℎ𝑘 + 1
𝑡 = ⌊𝑙𝑜𝑔3 𝑛⌋ − 1

Here we are using notation for the floor function ⌊𝑥⌋ means the largest integer ≤ 𝑥.
This results in a sequence 1, 4, 13, 40, 121.... You stop computing values in the sequence when 𝑡 = 𝑙𝑜𝑔3 𝑛 − 1. (So
for n=50,000, you should have about 9-10 intervals.)
For completeness, we note that 𝑙𝑜𝑔3 𝑛 must be sufficiently large (and > 2) for this method to work. Our code ensures
this by taking the maximum of 𝑙𝑜𝑔3 𝑛 and 1.
Knuth also suggests:

ℎ0 = 1
ℎ𝑘+1 = 2ℎ𝑘 + 1
𝑡 = ⌊𝑙𝑜𝑔2 𝑛⌋ − 1

This results in a sequence 1, 3, 7, 15, 31....

Here is the improvement to our naive method that dynamically calculates the intervals based on the first suggestion of
Knuth:
Shell sort is a complex sorting algorithm to make “work well”, which is why it is not seen often in practice. It is,
however, making a bit of a comeback in embedded systems.
We nevertheless think it is a very cool algorithm to have heard of as a computer science student and think it has promise
in a number of situations, especially in systems where there are limits on available memory (e.g. embedded systems).

10.4.6 Quicksort a.k.a. Partition Sort

This sort is a more advanced example that uses recursion. We list it because it is one of the best sorts for random
data, having an average time behavior of 𝑂(𝑁 log 𝑁 ). Quicksort is a rather interesting case. While it has an excellent
average behavior, it has a worst case performance of 𝑂(𝑁 2 ).
Though the initial call is to sort an entire array, this is accomplished by dealing with sections, so the main work is done
in the version with the two extra parameters, giving the lowest and highest index considered.
It picks an arbitrary element as pivot, and then swaps elements with values above and below the pivot until the part of
the array being processed is in three sections:
1. elements <= pivot;
2. possibly some of the elements equal to pivot
3. elements >= pivot.
Though sections 1 and 3 are not sorted, there are no inversions between them, so only the smaller sections 1 and 3
need to be sorted separately, and only then if they have at least two elements. They can be sorted by calling the
same function, just with a smaller range of indices to deal with in each case. These recursive calls stop when a part it
reduced to one element.

10.4.7 Random Data Generation

Now it is time to talk about how we are going to check the performance in a real-world situation. We’re going to start
by modeling the situation here the data are in random order.

10.4. Sorting Algorithms 141

Introduction to Computer Science in C#, Release 2.0

The following code generates a random array:

There are a few things to note in this code:
1. We use the random number generator option to include a seed. Random numbers aren’t truly random. The
particular sequence is just determined by a seed. The simplest way to create a Random object uses a seed taken
from the system clock.
2. Because the sorting algorithms modify the data that are passed to it, we need to have a way of regenerating the
sequence. (We could also copy the data, but it is kind of a waste of memory.)
3. In order to regenerate a particular example, we actually need the random sequence to be consistent, so we know
that each of the sorting algorithms is being tested using the same random data. Hence we specify the same seed
each time.

10.4.8 Timing

In this code, we are using another class from the .Net framework/library.
We need the ability to time the various sorting algorithms. Luckily, .Net gives us a way of doing so through its
Stopwatch class. This class supports methods that you would expect if you’ve ever used a stopwatch (the kind
found in sports):
• Reset: Resets the elapsed time to zero. We need this so we can use the same Stopwatch for each sorting
algorithm.
• Start: Starts the stopwatch. Will keep recording time until stopped.
• Stop: Stops the stopwatch.
• ElapsedMilliseconds: Not really a method but a property (like a variable). We’ll use this to get the total time
that has elapsed between pairs of Start/Stop events in milliseconds.
So let’s take a look at how we compare the sorting algorithms by looking at the Main() method’s code. As this code
is fairly lengthy, we’re going to look at parts of it. The Main() method should be thought of as an experiment that
tests the performance of each of the sorting algorithms.
The variables declared here are to set up the apparatus:
• arraySize: The size of the array where we wish to test the performance. We will use this to create an array
with arraySize random values.
• randomSeed: This allows the user to vary the seed that is used to create the random array. We often want
to do this to determine whether our performance results are stable when run a large number of times with
different distributions. We won’t go into too much detail here but consider it an important part of building good
performance benchmarks.
• watch: The stopwatch object we’re using to do the timings of all experiments.
This code is designed so we can accept the parameters arraySize and randomSeed from the command line
or by prompting the user. When programmers design benchmarks, they often try to make it possible to run them
with minimal user interaction. For the purposes of teaching, we wanted to make it possible to run it with or without
command-line parameters.
This code fragment is actually replicated a few times in the actual Main() method (to run each of the different sorting
algorithms). Essentially, we do the following for each of the sorting algorithms we want to benchmark:
1. Create the random array of data.
2. Reset the Stopwatch object to zero.
3. Start the Stopwatch.

142 Chapter 10. Arrays

 Introduction to Computer Science in C#, Release 2.0

4. Run the sorting algorithm of interest (here IntArrayBubbleSort()). In the rest of the Main() code, we
change this line to call the function for each of the other sorting algorithms.
5. Stop the Stopwatch and get the elapsed time (watch.Elapsed).
6. Print the performance results.
When you get watch.ElapsedMilliseconds, this gives you an integer (long) number of milliseconds (thou-
sandths of a second).
You can find this code in sorting/sorting.cs.

10.4.9 Running the Code

Here’s the output of a trial run on one of our computers. The results will vary depending on your computer’s CPU,
among other factors.
bin/Debug$ mono Sorting.exe 1000 12
Quick Sort: 0.000
Naive Shell Sort: 0.000
Better Shell Sort: 0.000
Insertion Sort: 0.001
Selection Sort: 0.002
Bubble Sort: 0.003
bin/Debug$ mono Sorting.exe 1000 55
Quick Sort: 0.000
Naive Shell Sort: 0.000
Better Shell Sort: 0.000
Insertion Sort: 0.001
Selection Sort: 0.002
Bubble Sort: 0.003
bin/Debug$ mono Sorting.exe 10000 2
Quick Sort: 0.001
Naive Shell Sort: 0.019
Better Shell Sort: 0.002
Insertion Sort: 0.134
Selection Sort: 0.174
Bubble Sort: 0.321
bin/Debug$ mono Sorting.exe 50000 2
Quick Sort: 0.006
Naive Shell Sort: 0.441
Better Shell Sort: 0.015
Insertion Sort: 3.239
Selection Sort: 4.172
Bubble Sort: 8.028
bin/Debug$ mono Sorting.exe 100000 2
Quick Sort: 0.014
Naive Shell Sort: 1.794
Better Shell Sort: 0.034
Insertion Sort: 13.158
Selection Sort: 16.736
Bubble Sort: 31.334

At least based on randomly-generated arrays, the performance can be summarized as follows:

• Bubble Sort is rather unimpressive as expected. In fact, this algorithm is never used in practice but is of historical
interest. Like the brute-force style of searching, it does way too much work to come up with the right answer!
• Selection Sort and Insertion Sort are also rather unimpressive on their own. Even though Selection Sort can in
theory do a lot less data movement, it must make a large number of comparisons to find the minimum value to

10.4. Sorting Algorithms 143

Introduction to Computer Science in C#, Release 2.0

be moved. Again it is way too much work. Insertion Sort, while unimpressive, fares a bit better and turns out to
be a nice building block (if modified) for the Shell Sort. Varying the interval size drastically reduces the amount
of data movement (and the distance it has to move).
• Shell Sort does rather well, especially when we pick the right intervals. In practice, the intervals also need to
be adjusted based on the size of the array, which is what we do as larger array sizes are considered. This is no
trivial task but a great deal of work has already been done in the past to determine functions that generate good
intervals.
• The Quicksort is generally fastest. It is by far the most commonly used sorting algorithm. Yet there are signs
that Shell sort is making a comeback in embedded systems, because it concise to code and is still quite fast. See
[WikipediaShellSort], where it is mentioned that the [uClibc] library makes use of Shell sort in its qsort()
implementation, rather than implementing the library sort with the more common quicksort.

10.5 Binary Searching

Binary search is an improvement over linear searching that works only if the data in the array are sorted beforehand.
Suppose we have the following array data shown under the array indices:
10 20 30 40 50 60 70 80 90 100 115 125 135 145 155 178 198

If we are looking for a number, say, 115, here is a visual on how we might go about it. We start with the indices over
the data being considered. Here min and max are the smallest and largest index to still consider:
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
10 20 30 40 50 60 70 80 90 100 115 125 135 145 155 178 198
min=0 max=16 mid=8
100 115 125 135 145 155 178 198
min=9 max=16 mid=12
100 115 125
min=9 max=11 mid=10
Item 115 found at position 10

Binary search works by keeping track of the midpoint (mid) and the minimum (min) and maximum (max) positions
where the item might be.
Let’s see how we might search for the value 115.
• We start by testing the data at position 8. 115 is greater than the value at position 8 (100), so we assume that the
value must be somewhere between positions 9 and 16.
• In the second pass, we test the data at position 12 (the midpoint between 9 and 16). 115 is less than the value at
position 12, so we assume that the value must be somewhere between positions 9 and 11.
• In the last pass, we test the value at position 10. The value 115 is at this position, so we’re done.
So binary search (as its name might suggest) works by dividing the interval to be searched during each pass in half.
If you think about how it’s working here with 17 items. Because there is integer division here, the interval will not
always be precisely half. it is the floor of dividing by 2 (integer division, that is).
You can see that the above determined the item within 3 steps. To reduce to one element to consider, it could be 4 or
5 steps. Note that 4 < 𝑙𝑜𝑔2 17 < 5.
Now that we’ve seen how the method works, here is the code that does the work:
Here’s a quick explanation, because it largely follows from the above explanation.
• Line 5-6: Initially item could be anywhere in the array, so minimum is at position 0 and maximum is at position
N-1 (data.Length - 1).

144 Chapter 10. Arrays

 Introduction to Computer Science in C#, Release 2.0

• The loop to make repeated passes over the array begins on line 7. We use a bottom-tested do-while loop, because
we know that we need to enter this loop at least once, no matter what, to determine whether the item is in the
middle or not.
• Line 8 does just what we expect: It calculates the median position (mid).
• It is always possible that we’ve found the item, which is what we test on line 9, and return with our answer if
we found it.
• Lines 11-14: If not, we continue. If the item is greater than the value at this mid position, we know it is in the
“upper half”. Otherwise, it’s in the “lower half”.
• Line 15: We can only continue searching if there is some data left to consider (min <= max).
• Line 16: Otherwise the binary search loop terminates, and we return -1 (to indicate not found). The -1 value is a
commonly-returned indicator of failure in search operations (especially on arrays, lists, and strings), so we use
this mostly out of respect for tradition. It makes particular sense, because -1 is not within the index set of the
array (which starts at 0 in C# and ends at data.Length - 1.
Similar to linear searching, we provide a main program that tests it out. The whole code is in bi-
nary_searching/binary_searching.cs. It uses an elaboration of binary search that prints out the steps visually, as in
the introduction to this section. It also references functions from the example projects searching and sorting.

10.5.1 Elaborated Binary Search Exercise

Even if you do not find item in a binary search, it is sometimes useful to know where item lies in relation to the
array elements. It could be before the first element, in between two elements, or after the last element. Suppose N is
the (positive) array length. Instead of just returning -1 if item is not in the array, return
• -1 if item < data[0]
• -(k+1) if data[k-1] < item < data[k]
• -(N+1) if data[N-1] < item
Modify binary_searching/binary_searching.cs into binary_searching2.cs so this extra information is returned
(and indicated clearly in the main testing program). This should not require an extra loop or test in the binary search.

10.6 Lab: Arrays

10.6.1 Overview

In this lab, we’ll learn how to work with arrays. Arrays are fundamental to computer science, especially when it comes
to formulating various algorithms. We’ve already learned a bit about arrays through the string data type. In many
ways, a character string reveals the secrets of arrays:
• each element of a string is a common type (char)
• we can use indexing to find any given character, e.g. s[i] gives us the character at position i.
• we know that the string has a finite length, e.g. s.Length.
So you’ve already learned the concept. But what if we want to have arrays that can hold different kinds of data, for
example, integer or floating point data? That’s why we need this lab.

10.6. Lab: Arrays 145

Introduction to Computer Science in C#, Release 2.0

10.6.2 Goals

In this lab, we’re going to practice:

• creating arrays that hold numerical data
• populating an array with data
• using the tools of loops and decisions to do something interesting with the data
• printing the data

10.6.3 Tasks

Copy the example file array_lab_stub/array_lab.cs to a new project of yours. Complete the body of a function for each
main part, and call each function in Main several times with actual parameters chosen to test it well. To label your
illustrations, make liberal use of the first function, PrintNums, to display and label inputs and outputs. Where several
tests are appropriate for the same function, you might want to write a separate testing function that prints and labels
inputs, passes the data on to the function being tested, and prints results.
Recall that you can declare an array in two ways:
int[] myArray1 = new int[10];
int[] myArray2 = { 7, 7, 3, 5, 5, 5, 1, 2, 1, 2 };

The first declaration creates an array initialized to all zeroes. The second creates an array where the elements are taken
from the bracketed list of values. The second will be convenient to set up tests for this lab.
1. Complete and test the function with documentation and heading:
This will be handy for labeling later tests. Note that you end on the same line, but a later label can start with \n
to advance to the next line.
2. Complete and test the function with documentation and heading:
This will allow user tests. The earlier input utility functions are included at the end of the class.
3. Complete and test the function with documentation and heading:
4. Complete and test the function with documentation and heading:
5. Complete and test the function with documentation and heading:
To test this out, you’ll need to declare and initialize the arrays to be added. You’ll also need to declare a third
array to hold the results. Make sure that the arrays all have the same dimensionality before proceeding.
This section is a warm-up for the next one. It is not required if you do the next one:
6. Complete and test the function with documentation and heading:
See how this is different from the previous part!
7. Complete and test the function with documentation and heading:
This has some pitfalls. You will need more tests that the ones in the documentation! You can code this with a
“short-circuit” loop. What do you need to find to be immediately sure you know the answer?
8. 20 % extra credit: Complete and test the function with documentation and heading:
9. 20 % extra credit: Complete and test the function with documentation and heading:
10. 20 % extra credit: Given two arrays, a and b that represent vectors. Write a function that computes the vector
dot product of these two floating point arrays. The vector dot product (in mathematics) is defined as the sum of
a[i] * b[i] (for all i). Here’s an example of how it should work:

146 Chapter 10. Arrays

 Introduction to Computer Science in C#, Release 2.0

double[] a = new double[] { 1.0, 2.0, 3.0 };

double[] b = new double[] { 4.0, 2.0, -1.0 };

double dotProduct = VectorDotProduct(a, b);

Console.WriteLine("The dot product is {0}", dotProduct);

// Should print 1.0 * 4.0 + 2.0 * 2.0 + 3.0 * -1.0 = 5.0

From here on, create your own headings.

11. 20 % extra credit: Suppose we have loaded an array with the digits of an integer, where the highest power is
kept in position 0, next highest in position 1, and so on. The ones position is always at position array.Length - 1:
int[] digits = { 1, 9, 6, 7 };

Without showing you the code, here is how you would convert a number from its digits to an integer:
num = 0
num = 10 * 0 + 1 = 1
num = 10 * 10 + 9 = 19
num = 10 * 19 + 6 = 196
num = 10 * 196 + 7 = 1967
done!

Write a function that converts the array of digits representing a base 10 number to its int value (or for really
long integers, you are encouraged to use a long data type). Note that we only allow single digit numbers to be
placed in the array, so negative numbers are not addressed.
12. 20 % extra credit: Suppose that we not only have the digits but also the base that in which the number is
represented. (The base can be at most 10 if it uses only digits for place value.) Write a function (or revise the
previous solution) to return the int or long represented. For example if {1, 0, 0, 1, 1} represents a base 2 number,
19 is returned.

10.7 Lab: Performance

In Sorting Algorithms we took advantage of a few ideas to show how to do basic benchmarking to compare the various
approaches.
• using randomly-generated data
• making sure each algorithm is working with the same data
• making sure that we try a range of sizes to observe the effects of scaling
• using a timer with sufficiently high resolution (the Stopwatch gives us measurements in milliseconds).
In this lab, you get your chance to learn a bit more about performance by comparing searches. The art of benchmarking
is something that is easy to learn but takes a lifetime to master (to borrow a phrase from the famous Othello board
game).
Most of the algorithms we cover in introductory courses tend to be polynomial in nature. That is, the execution time
can be expressed as a polynomial function of the size of the data size 𝑛. Examples include but are not limited to:
• 𝑂(1) is constant time, characterized by a calculation with a limited number of steps.
• 𝑂(𝑛) is linear time; often characterized by a single loop
• 𝑂(𝑛2 ) is the time squared; often characterized by a nested loop

10.7. Lab: Performance 147

Introduction to Computer Science in C#, Release 2.0

• 𝑂(𝑙𝑜𝑔 𝑛) is logarithmic (base 2) time; often characterized by a loop that repeatedly divides its work in half. The
binary search is a well-known example.
• 𝑂(𝑛 𝑙𝑜𝑔 𝑛) is an example of a hybrid. Perhaps there is an outer loop that is linear and an inner loop that is
logarithmic.
And there are way more than these shown here. As you progress in computing, you’ll come to know and appreciate
these in greater detail.
In this lab, we’re going to look at a few different data structures and methods that perform searches on them and do
empirical analysis to get an idea of how well each combination works. Contrasted with other labs where you had to
write a lot of code, we’re going to give you some code to do all of the needed work but ask you to do the actual analysis
and produce a basic table.

10.7.1 The Experiments

We’re going to measure the performance of data structures we have been learning about (and will learn about, for lists
and sets). For this lab, we’ll focus on:
• Integer arrays using Linear Searching and Binary Searching
• Lists of integers with linear searching
• Sets of integers; checking if an item is contained in the set
In the interest of fairness, we are only going to look at the time it takes to perform the various search operations.
We’re not going to count the time to randomly-generate the data and actually build the data structure. The reasoning
is straightforward. We’re interested in the search time, which is completely independent of other aspects that may be
at play. We’re not at all saying that the other aspects are unimportant but want to keep the assignment focus on search.
The experimental apparatus that we are constructing will do the following for each of the cases:
• create the data structure (e.g. new array, new list, new set)
• use a random seed seed, initialize a random generator that will generate n values.
• insert the random values into the data structure . For the case of sets, which eliminate duplicates, it is entirely
possible you will end up with a tiny fraction of a percent fewer than n values.
• to measure the performance of any given search method, we need to perform a significant number of lookups
(based on numbers in the random sequence) to ensure that we get an accurate idea of the average lookup time
in practice. We’ll call this parameter, rep. We will spread out the values looked for by checking data elements
that have indices at a regular interval throughout the array. The separation is m = n/rep when rep < n. We
wrap around if rep > n.
• We’ll start a Stopwatch just before entering the loop to perform the lookups.

10.7.2 Starter Project

To make your life easier, we have put together a project that refers to all the code for all of the experiments you need
to run. (That’s right, we’re giving you the code for the experiments, but you’re going to write some code to run the
various experiments and then run for varying sizes of n.) The stub file is performance_lab_stub/performance_lab.cs.
Recreate example project performance_lab_stub in your solution as performance_lab, so you have your own copy
to modify. You will need to copy into the lab project the file sorting.cs from the example sorting project, and
the file searching.cs from the linear searching project, and the file binary_searching.cs from the binary
searching project. (An alternative is to recreate their whole projects, and reference them from the lab project.)
Here is the code for the first experiment, to test the performance of linear searching on integer arrays:

148 Chapter 10. Arrays

 Introduction to Computer Science in C#, Release 2.0

Let’s take a quick look at how this experiment is constructed. We’ll also take a look at the other experiments but these
will likely be presented in a bit less detail, except to highlight the differences:
• On line 3, we create a Stopwatch instance. We’ll be using this to do the timing.
• On lines 4-5, we are creating the data to be searched. Because we have already written this code in our sorting
algorithms examples, we can refer to the Sorting class code in sorting.cs, as long as you made the lab
project able to reference it. We use the Sorting class name to access the method IntArrayGenerate()
within this class. We also take advantage of this in the other experiments.
• Line 6 converts the number of repetitions into the increment in index values for each time.
• Line 7 resets the stopwatch. It is not technically required; however, we tend to be in the habit of doing it, because
we sometimes reuse the same stopwatch and want to make sure it is completely zeroed out. A call to Reset()
ensures it is zero.
• Line 8 actually starts the stopwatch. We are starting here as opposed to before line 3, because the random data
generation has nothing to do with the actual searching of the array data structure.
• Lines 10 through 12 are searching rep times for an item already known to be in the array.
• Line 13 stops the stopwatch.
• Line 14 returns the elapsed time in milliseconds between the Start() and Stop() method calls, which
reflects the actual time of the experiment.
Each of the other experiments is constructed similarly. For linear search and binary search we use the methods created
earlier. For the lists and the set we use the built-in Contains method to search. The list and set are directly initialized
in their constructors from the array data. (More on that in later chapters.)
You need to fill in the Main method:
1. Write the code to parse command line args for the parameters rep and any number of values for n. For
instance:
mono PerformanceLab.exe 50 1000 10000 100000
would generate the table shown below for 50 repetitions for each of the values of n: 1000, 10000, and 100000.
2. Write the code to run each of the experiments for rep and a given value of n.
3. Iterate through the values of n and print a nice right-justified table, something like the following, with the
number of seconds calculated. Experiment and adjust the repetitions to get perceptible values. Our choice of 50
may not be appropriate with these n values.
n rep linear list binary set
1000 50 ????.??? ????.??? ??.??? ??.???
10000 50 ????.??? ????.??? ??.??? ??.???
100000 50 ????.??? ????.??? ??.??? ??.???

The table would be longer if more values of n were entered on the command line. Note that the experi-
ments return times in milliseconds, (1/1000 of a second) while the table should print times in seconds.
Because the range of speeds is so enormous, make an accommodation with the slow linear ver-
sions: If rep >= 100 and (long)n*rep >= 100000000, then, for the linear and list columns
only, time with rep2 = rep/100 instead of rep, and then compensate by multiplying the time by
(double)rep/rep2 to produce the final table value. (This multiplier is not just 100, since the integer
division creating rep2 may not be exact.)
Once again, you are encouraged to develop this is steps, for example
1. Make sure you can parse the command line parameters.
2. Print out one test for one rep, n.

10.7. Lab: Performance 149

Introduction to Computer Science in C#, Release 2.0

3. Print out the results for all tests for one rep, n
4. Make the printing be formatted for the columns
5. The full program, for all n.

10.8 Multi-dimensional Arrays

10.8.1 Rectangular Arrays (Two Dimensional)

You should be familiar with one dimensional arrays. The data in arrays may be any type. While a one dimensional
array works for a sequence of data, we need something more for a two dimensional table, where data values vary over
both row and column.
If we have a table of integers, for instance with three rows and four columns:
2 4 7 55
3 1 8 10
6 0 49 12

We could declare an array variable of the right size as

int[,] table = new int[3, 4];

Multiple indices are separated by commas inside the square brackets. In declaring an array type, no indices are
included so the [,] indicates a two dimensional array. Where the new object is being created, the values inside the
square brackets give each dimension.
In general the notation for a two dimensional array declaration is:
type [, ] variableName
and to create a new array with default values:
new type [ intExpression1, intExpression2 ]
where the expressions evaluate to integers for the dimensions.
To assign the 8 in the table above, consider that it is in the second row in normal counting, but we start array indices
at 0, so there are rows 0, 1, and 2, and the 8 has row index 1. Again starting with index 0 for columns, the 8 is at index
2. We can assign a value to that position with
table[1, 2] = 8;

In fact a two dimensional array just needs two indices that could mean anything. For instance, it was just the standard
convention that we called the left index the row. They could have been switched everywhere, and assume the notation
table[column, row]:
int[,] table = new int[4, 3];
table[2, 1] = 8;

but we will stick with the original [row, column] model.

Data indexed by more than two integer indices can be stored in a higher dimension array, with more indices between
the square braces. We will only consider two dimensional arrays in the examples here.
A shorthand for initializing all the data in the table, analogous to initializing one dimensional arrays is:
int[, ] table = { {2, 4, 7, 55},
{3, 1, 8, 10},
{6, 0, 49, 12} };

150 Chapter 10. Arrays

 Introduction to Computer Science in C#, Release 2.0

All rows must be the same length when using this notation.
Often two dimensional arrays, like one dimensional arrays, are processed in loops. Multiple dimension arrays are
often processed in nested loops. We could print out this table using columns 5 spaces wide with the code:
If we wanted to make a more general function out of that code, we have a problem: the number of rows and columns
were literal values that we knew. We need something more general. For one dimensional arrays we had the Length
property, but now there are more than one lengths!
The following csharp sequence illustrates the syntax needed:
csharp> int[, ] table = { {2, 4, 7, 55},
> {3, 1, 8, 10},
> {6, 0, 49, 12} };
csharp> table.Length;
12
csharp> table.GetLength(0);
3
csharp> table.GetLength(1);
4
csharp> foreach (int n in table) {
> Console.WriteLine(n);
> }
2
4
7
55
3
1
8
10
6
0
49
12

Note:
• The meaning for the Length property, is now the total number of elements, (3)(4) = 12.
• The separate method GetLength is needed to find the number of rows and columns. The entries in the list
of array indices in the multi-dimensional array notation are themselves indexed to provide the GetLength
method parameter for each dimension. In this case 0 gives the row length (left index of the array notation), and
1 gives the column length (right index of the array notation).
• The foreach statement behavior is consistent with the Length property, giving all the elements, row by row.
More generally, the rightmost indices vary most rapidly as the foreach statement iterates through all elements.
Just replacing 3 and 4 by table.GetLength(0) and table.GetLength(1) in the table printing code would
make it general.
A more elaborate table might include row and column sums:
2 4 7 55 | 68
3 1 8 10 | 22
6 0 49 12 | 67
---------------------
11 5 64 77 | 157

For example, the following function from example file print_table/print_table.cs, prints out a table of integers neatly,
including row and column sums. It illustrates a number of things. It shows the interplay between one and two

10.8. Multi-dimensional Arrays 151

Introduction to Computer Science in C#, Release 2.0

dimensional arrays, since the row and column sums are just one dimensional arrays.
Now that we are using arrays, we can easily look at the same collection of data repeatedly. It is possible to look at
the data one time to just see its maximum width, and then go through again and print data using a column width that
is just large enough for the longest numbers. When looking through the data for string lengths, the row and column
structure is not important, so the code illustrates foreach loops to chug through all the data. We use the earlier trick
in Modular Multiplication Table, creating a custom format string to make columns the right width.
The code refers once to the earlier StringOfReps in Lab: Loops for the row of dashes setting off the column sums:

Row and Column Numbering Exercise

Write a function that sets the values in a given rectangular array to 10 * (row index +1) + column index + 1, with the
normal row and column indices, starting from 0. For example an array with two rows and five columns would end up
with values below. Your method should set the values in the array, not print them out:
11 12 13 14 15
21 22 23 24 25

If there are no more than 9 rows or columns, this display gives row and column numbers neatly for the normal human
counting system, starting from 1.

Varying Column Width Exercise

Copy the project file print_table/print_table.cs to a file print_varying_width_table.cs in a projet of yours.

Edit it so that each column is only as wide as it needs to be: the width for the widest entry in that column. The earlier
data would now print as:
2 4 7 55 | 68
3 1 8 10 | 22
6 0 49 12 | 67
----------------
11 5 64 77 | 157

Hint: Create an array of widths and an array of format strings.

10.8.2 Advanced topic: Array of Arrays

It is also possible to index a collection of rows of data, where the lengths of the rows vary. This can be done with the
less friendly syntax for an array whose elements have array type. Here is code to make a doubly indexed “triangular”
collection. Each row must be separately created as a new array:
int[][] tri = new int[4][]; //create four null int[] elements
for (int i = 0; i < tri.Length; i++) { // Length 4 (rows)
tri[i] = new int[i+1]; // each row is a different length
}

Individual entries would be referred to with double square braces, not with indices separated by commas between one
set of braces. With tri constructed as above:
• tri[3, 3] would given a compiler error: no changing [][] to [, ].
• tri[3][3] would refer to an element.
• tri[1][3] would given a run-time out-of-bounds error. since tri[1] is an array of length 2.

152 Chapter 10. Arrays

 Introduction to Computer Science in C#, Release 2.0

Stick with our regular rectangular arrays, with commas between indices, unless you have a very special reason to use
arrays of arrays.

10.8. Multi-dimensional Arrays 153

Introduction to Computer Science in C#, Release 2.0

154 Chapter 10. Arrays

 CHAPTER

ELEVEN

LISTS

11.1 List Syntax

Arrays are fine if you know ahead of time how long your sequence of items is. Then you create your array with that
length, and you are all set.
If you want a variable sized container, you are likely to want a List. As with arrays, you might want a collection of
any type. Unfortunately, you cannot use the simple notation of arrays to specify the type of element in a List. Arrays
are built into the language, and they have their own special syntax. Lists are handled in the library of types provided
by C# from the .Net framework. There are all sorts of situations where you might want a general idea to have a version
for each of many kinds of objects. There is not a new syntax for each one.

11.1.1 Generics

Instead .Net 4.0 introduced one new form of syntax that can apply to all sorts of classes, generics.
The type for a list of strings is
List<string>

The type for an int list is

List<int>

In general the new generic syntax allows a type (or several, comma separated) in angle brackets after a class name.
Lists are an example that depends on just one included type. We will see more shortly.
There is a namespace for the generics for collections, including List: System.Collections.Generic.
We will use several generic library classes, though we will not write the definitions of new generic classes ourselves.

11.1.2 List Constructors and Methods

We can play with some List methods in csharp. Note that csharp informally displays the value of a List with a
list of elements inside braces. This is not a legal way to assign values to lists. Nor is it the unhelpful way C# would
print out a string representation of a List with Console.WriteLine. The blocks below are all from one csharp
session, with our comments breaking up the sequence.
With the no-parameter constructor, the List is empty to start:
csharp> List<string> words = new List<string>();
csharp> words;
{ }

155
Introduction to Computer Science in C#, Release 2.0

csharp> words.Count;
0

You can add elements, and keep count with the Count property as the size changes:
csharp> words.Add("up");
csharp> words;
{ "up" }
csharp> words.Add("down");
csharp> words;
{ "up", "down" }
csharp> words.Add("over");
csharp> words;
{ "up", "down", "over" }
csharp> words.Count;
3

You can reference and change elements by index, like with arrays:
csharp> words[0];
"up"
csharp> words[2];
"over"
csharp> words[2] = "in";
csharp> words;
{ "up", "down", "in" }

You can use foreach like with arrays or other sequences:

csharp> foreach (string s in words) {
> Console.WriteLine(s.ToUpper());
> }
UP
DOWN
ON

Compare Remove, which finds the first matching element and removes it, and RemoveAt, which removes the element
at a specified index. Remove returns whether the List has been changed:
csharp> words.Remove("down");
true
csharp> words;
{ "up", "in" }
csharp> words.Remove("around"); // no change
false
csharp> words.Add("out");
csharp> words.Add("on");
csharp> words;
{ "up", "in", "out", "on" }
csharp> words.RemoveAt(2); // "out" is at index 2
csharp> words;
{ "up", "in", "on" }

Removing does not leave a “hole” in the List: The list closes up, so the index decreases for the elements after the
removed one:
csharp> words[2];
"on"
csharp> words.Count;
3

156 Chapter 11. Lists

 Introduction to Computer Science in C#, Release 2.0

You can check for membership in a List with Contains:

csharp> words.Contains("in");
true
csharp> words.Contains("into");
false

You can also remove all elements at once:

csharp> words.Clear();
csharp> words.Count;
0

Here is a List containing int elements. Though more verbose than for an array, you can initialize a List with another
collection, including an anonymous array, specified with an explicit list in braces:
csharp> List<int> nums = new List<int>(new[]{5, 3, 7, 4});
csharp> nums;
{ 5, 3, 7, 4 }

We have been using the explicit declaration syntax, but generic types tend to get long, so var is handy with them:
var stuff = new List<string>();

When initializing a generic object, you still need to remember both the angle braces around the type and the parentheses
for the parameter list after that.
An aside on the Remove method: It both causes a side effect, changing the list, and it returns a value. If a function
returns a value, we typically use the function call as an expression in a larger statement. This is not necessary, as
described in Not using Return Values. In that section we discussed the mistake of not using return values. The
Remove method illustrates that this is not always a mistake: If you just want the side effect, trying to remove an
element, whether or not it is in the list, then there is no need to check for the return value. This complete C# statement
is fine:
someList.Remove(element);

You should generally think carefully before defining a function that both has a side effect and a return value. Most
functions that return a value do not have a side effect. If you see a function used in the normal way as an expression,
it is easy to forget that it was also producing some side effect.

11.1.3 Interactive List Example

Lists are handy when you do not know how much data there will be. A simple example would be reading in lines from
the user interactively:
// Return a List of lines entered by the user in response
// to the prompt. Lines in the List will be nonempty, since an
// empty line terminates the input.
List<string> ReadLines(string prompt)
{
List<string> lines = new List<string>();
Console.WriteLine(prompt);
Console.WriteLine("An empty line terminates input.");
string line = Console.ReadLine();
while (line.Length > 0) {
lines.Add(line);
line = Console.ReadLine();
}

11.1. List Syntax 157

Introduction to Computer Science in C#, Release 2.0

return lines;
}

11.2 .Net Library (API)

This book can only introduce so many classes and methods from the C# library. You should browse the MSDN .Net
Framework Class Library’s online documentation.
http://msdn.microsoft.com/en-us/library/gg145045.aspx
We mostly deal with classes in the namespaces System, System.IO, and System.Collections.Generic, and you can drill
down to them.
One complication is that variations on these classes and methods are included for several Microsoft languages. Under
the Syntax heading, make sure the C# tab is selected.
For example, you can click in the left column on System.Collections Namespaces, and then Sys-
tem.Collections.Generic, and then, for example, List(T). (In C# that is List<T>.)
The summary section separates constructors, properties, and methods. When you see one of these with promise, click
on it to get the full details. For example, click on the first method in the Methods section, Add, or something new, like
IndexOf, or Reverse, or Sort....
Classes also can be classified in several ways for browsing:
• Those you will want to be fairly familiar with pretty soon: string, List, Dictionary
• Those that might be useful, that you should be at least aware of.
• Those that may be useful eventually, but are not worth your time now.
You will also find methods in various categories as you browse:
• Methods that make sense and are useful right away
• Methods that take a little reading to absorb
• Features that we have yet to discuss
• Features that are well beyond what we have talked about - ask or wait or read a lot.

158 Chapter 11. Lists

 CHAPTER

TWELVE

DICTIONARIES

12.1 Dictionary Syntax

We have explored several ways of storing a collection of the same type of data:
• arrays: built-in syntax, unchanging size of the collection
• List: generic class type, allows the size of the collection to grow
Both approaches allow reference to data elements using a numerical index between square brackets, as in words[i].
The index provides an order for the elements, but there is no meaning to the index beyond the sequence order.
Often, we want to look up data based on a more meaningful key, as in a dictionary: given a word, you can look up the
definition.
C# uses the type name Dictionary, but with greater generality than in nontechnical use. In a regular dictionary,
you start with a word, and look up the definition. The generalization is to have some piece of data that leads you to (or
maps to) another piece of data. The computer science jargon is that a key leads you to a value. In a normal dictionary,
these are both likely to be strings, but in the C# generalization, the possible types of key and value are much more
extensive. Hence the generic Dictionary type requires you to specify both a type for the key and a type for the
value.
We can initialize an English-Spanish dictionary e2sp with
Dictionary<string, string> e2sp = new Dictionary<string, string>();

That is quite a mouthful! The C# var syntax is handy to shorten it:

var e2sp = new Dictionary<string, string>();

The general generic type syntax is

Dictionary< keyType, valueType >
If you are counting the number of repetitions of words in a document, you are likely to want a Dictionary mapping
each word to its number of repetitions so far:
var wordCount = new Dictionary<string, int>();

If your friends all have a personal list of phone numbers, you might look them up with a dictionary with a string name
for the key and a List of personal phone number strings for the value. The type could be Dictionary<string,
List<string>>. This example illustrates how one generic type can be built on another.
There is no restriction on the value type. There is one important technical restriction on the key type: it should be
immutable. This has to do with the implementation referenced in Dictionary Efficiency.
Similar to an array or List, you can assign and reference elements of a Dictionary, using square bracket notation.
The difference is that the reference is through a key, not a sequential index, as in:

159
Introduction to Computer Science in C#, Release 2.0

e2sp["one"] = "uno";
e2sp["two"] = "dos";

Csharp displays dictionaries in its own special form, as a sequence of pairs {key, value}. Again, this is special to
csharp. Here is a longer csharp sequence, broken up with our comments:
csharp> Dictionary<string, string> e2sp = new Dictionary<string, string>();
csharp> e2sp;
{}
csharp> e2sp["one"] = "uno";
csharp> e2sp["two"] = "dos";
csharp> e2sp["three"] = "tres";
csharp> e2sp.Count;
3
csharp> e2sp;
{{ "one", "uno" }, { "two", "dos" }, { "three", "tres" }}
csharp> Console.WriteLine("{0}, {1}, {2}...", e2sp["one"],
> e2sp["two"], e2sp["three"]);
uno, dos, tres...

If you want to iterate through a whole Dictionary, you will want the syntax below, with foreach and the property
Keys:
csharp> foreach (string s in e2sp.Keys) {
> Console.WriteLine(s);
> }
one
two
three

The documentation for Dictionary says that you cannot depend on the order of processing with foreach, though
the present implementation remembers the order in which keys were added.
It is often useful to know if a key is already in a Dictionary: Note the method ContainsKey:
csharp> e2sp.ContainsKey("seven");
false
csharp> e2sp.ContainsKey("three");
true

The method Remove takes a key as parameter. Like a List and other collections, a Dictionary has a Clear
method:
csharp> e2sp.Count;
3
csharp> e2sp.Remove("two");
true
csharp> e2sp.Count;
2
csharp> e2sp.Clear();
csharp> e2sp.Count;
0

12.2 Dictionary Efficiency

We could simulate the effect of a Dictionary pretty easily by keeping a List keys and a List values, in the same
order. We could find the entry with a specified key with:

160 Chapter 12. Dictionaries

 Introduction to Computer Science in C#, Release 2.0

int i = keys.IndexOf(key);
return values[i];

Searching though a List, however, take time proportional to the length of the List in general, linear order. Through
a clever implementation covered in data structures classes, a Dictionary uses a hash table to make the average
lookup time of constant order. A hash table depends on the keys being immutable.

12.3 Dictionary Examples

12.3.1 Sets

In the next section we will have an example making central use of a dictionary. It will also make use of a set.
The generic C# version is a HashSet, which models a mathematical set: a collection with no repetitions and no
defined order. We use a HashSet for the words to be ignored. We use a HashSet rather than a List because the
Contains method for a List has linear order, while the Contains method for a HashSet uses the same trick as
in a Dictionary to be of constant order on average.
Here is a csharp session using the type HashSet of strings. The Add method, like the Remove method for Lists,
returns true or false depending on whether the method changes the set:
csharp> var set = new HashSet<string>();
csharp> set;
{ }
csharp> set.Add("hi");
true
csharp> set;
{ "hi" }
csharp> set.Add("up");
true
csharp> set;
{ "hi", "up" }
csharp> set.Add("hi");
false
csharp> set;
{ "hi", "up" }
csharp> set.Contains("hi");
true
csharp> set.Contains("down");
false
csharp> var set2 = new HashSet<string>(new string[]{"a", "be", "see"});
csharp> set2;
{ "a", "be", "see" }

That lack of order for a HashSet means it cannot be indexed, but otherwise it has mostly the same methods and
constructors that have been discussed for a List, including Add and Contains and a constructor that takes a
collection as parameter.

12.3.2 Word Count Example

Counting the number of repetitions of words in a text provides a realistic example of using a Dictionary. With
each word that you find, you want to associate a number of repetitions. A complete program is in the example file
count_words/count_words.cs.
The central functions are excerpted below, and they also introduce some extra features from the .Net libraries.

12.3. Dictionary Examples 161

Introduction to Computer Science in C#, Release 2.0

This constructor pattern taking the elements of one collection and creating another collection, possibly of another type,
is used twice: first to create a HashSet from an array, and later to create a List from a HashSet. The latter is
needed so the List can be sorted in alphabetical order with its Sort method, used here for the first time. Our table
contains the words in alphabetical order.
Also used for the first time are two string methods: the pretty clearly named ToCharArray and another variation on
Split. An alternative to supplying a single character to split on, is to use a char array as parameter, and the string
is split at an occurrence of any of the characters in the array. This allows a split on all punctuation and special symbol
characters, as well as a blank.
We separate the processing into two functions, one calculating the dictionary, and one printing a table. To reduce the
amount of clutter in the Dictionary, the function GetCounts takes as a parameter a set of words to ignore.
Look at the code carefully, and look at the whole program that analyses the Gettysburg Address.

12.4 Lab: File Data and Collections

12.4.1 Goals for this lab:

• Read a text file.

• Work with loops.
• Work with a Dictionary and a List.
• Retrieve a random entry.

Overview

Copy project dict_lab_stub to your own project.

This lab provides a replacement file fake_help.cs for an improved project. The project still needs some additions in a
helper class.
Before we get there, open the comparison program fake_help_verbose/fake_help_verbose.cs and look at the methods
GetParagraphs() and GetDictionary(). All the strings for the responses are pre-coded for you there, but if
you were writing your own methods, it would be a pain. There is all the repetitious code to make multiline strings and
then to add to the List and Dictionary. This lab will provide simple versatile methods to fill a List<string> or a
Dictionary<string, string>: You only need you to write the string data itself into a text file, with the only
overhead being a few extra newlines. Minor further adaptations could save time later in your game project, too.
Look in your copy of fake_help.cs. It creates the List guessList and the Dictionary responses using more
general functions that you need to fill in. The stubs for these new versions are put in the class FileUtil for easy
reuse. Main calls these functions and chooses the files to read. The results will look the same as the original program
to the user, but the second version will be easier for a programmer to read and generalize: It will be easier in other
situations where you want lots of canned data in your program (like in a game you might write soon).
The stub should run as is (mostly saying things are not implemented). Test out your work at every stage!
You will need to complete very short versions of functions GetParagraphs and GetDictionary that have been
moved to file_util.cs and now take a StreamReader as parameter. The files that they read will contain the basic
data. You can look in the lab project at the first data file: help_not_defaults.txt, and the beginning is shown below:
You can see that it includes the data for the welcome and goodbye strings followed by all the data to go in the List
of random answers.

162 Chapter 12. Dictionaries

 Introduction to Computer Science in C#, Release 2.0

One complication is that many of these strings take up several lines, in what we call a paragraph. We follow a standard
convention for putting paragraphs into plain text: Put a blank line after a paragraph to mark its end. As you can see,
that is how help_not_defaults.txt is set up.

12.4.2 Steps

All of the additions you need to make are in bodies of function definitions in the class FileUtil. Look back to Main
in FakeAdvise to see how the functions from FileUtil are actually used: The StreamReader is set up to read
from the right file. The the FileUtil functions ReadParagraph, GetParagraphs, and GetDictionary
are used to provide the text data needed.

ReadParagraph

The first method to complete in file_util.cs is useful by itself and later for use in the GetParagraphs and
GetDictionary that you will complete. See the stub:
The first call to ReadParagraph, using the file illustrated above, should return the following (showing the escape
codes for the newlines):
"Welcome to We-Give-Answers!\nWhat do you have to say?\n"

and then the reader should be set to read the goodbye paragraph (the next time ReadParagraph is called).
To code, you can read lines one at a time, and append them to the part of the paragraph read so far. There is one
thing to watch out for: The ReadLine function throws away the following newline ("\n") in the input. You need to
preserve it, so be sure to explicitly add a newline, back onto your paragraph string after each nonempty line is added.
The returned paragraph should end with a single newline.
Throw away the empty line in the input after the paragraph. Make sure you stop after reading the empty line. It is very
important that you advance the reader to the right place, to be ready to read the next paragraph.
Be careful of a pitfall with files: You can only read a given chunk once: If you read again you get the next part.
This first short ReadParagraph function should actually be most of the code that you write for the lab! The program
is set up so you can immediately run the program and test ReadParagraph: It is called to read in the welcome string
and the goodbye string for the program, so if those come correctly to the screen, you can advance to the next two parts.

GetParagraphs

Since you have ReadParagraph at your disposal, you now only need to insert a few remaining lines of code to
complete the next method GetParagraphs, that reads to the end of the file, and likely processes more than one
paragraph.
Look again at help_not_defaults.txt, to see how the data is set up.
This lab requires very few lines of code. Be sure to read the examples and instructions carefully (several times). A lot
of ideas get packed into the few lines!
As a test after writing GetParagraphs, the random responses in the lab project program should work as the user
enters lines in the program.

GetDictionary

The last stub to complete in file_util.cs is GetDictionary. Its stub also takes a StreamReader as parameter. In
Main this function is called to read from help_not_responses.txt. Here are the first few lines:

12.4. Lab: File Data and Collections 163

Introduction to Computer Science in C#, Release 2.0

Here is the stub of the function to complete, reading such data:

When you complete this function, the program should behave like the earlier verbose version with the hard-coded data.
Be careful to distinguish the data file help_not_responses.txt from help_not_responses2.txt, used in the extra credit
option.
This should also be an extremely short amount of coding! Think of following through the data file, and get the
corresponding sequence of instructions to handle the data in the exact same sequence.
Show the program output to a TA (after the extra credit if you like).

Extra credit

1. (20%) Modify ReadParagragh so it also works if the paragraph ends at the end of the file, with no blank line
after it, or if the line after the paragraph only has whitespace characters. Both changes are good to bullet-proof
the code, since the added or removed whitespace is hard to see in print.
2. (20%) The crude word classification scheme would recognize “crash”, but not “crashed” or “crashes”. You
could make whole file entries for each key variation, repeating the value paragraph. A concise approach is to
use a data file like help_not_responses2.txt. Here are the first few lines:
The line that used to have one key now may have several blank-separated keys.
Here is how the documentation for GetDictionary should be changed:
Modify the lab project to use this file effectively: Find “help_not_responses.txt” on line 22 in Main. Change it
to “help_not_responses2.txt” (inserting ‘2’), so Main reads it.
In your test of the program, be sure to use several of the keys that apply to the same response, and show to your
TA.

164 Chapter 12. Dictionaries

 CHAPTER

THIRTEEN

CLASSES AND OBJECT-ORIENTED PROGRAMMING

13.1 A First Example of Class Instances: Rational

13.1.1 Making a Datatype

C# comes with lots of built-in datatypes, but not everything we might want to use. What if you want to use fractions,
which determine rational numbers?
You could always keep two integer variables, the numerator and denominator, but conceptually the main idea is that
of a (single) rational number. It just happens to have parts. There are lots of operations on rational numbers, and they
are operations on the entire fraction, as a unit.
It makes sense to think of a fraction or rational number as one entity. A way to do that is to create a class. We will
call our class Rational. This way we can have a single variable refer to a Rational object.
There is an alternate construction, that has most of the same properties, in Classes And Structs.
We have already considered built-in types with internal state, like Lists: Each List can contain different data, and the
internal data can be changed.
The idea of creating a new type of object opens new ground for managing data. Thus far we have stored data as local
variables, and we have called functions, with two ways to get information in and out of functions:
1. In through parameters and out through returned data.
2. Directly via the user: in through the keyboard and out to the screen.
We have stored and passed around built-in types of object using this model.
Now we have alternatives for storing and accessing data in the methods within a new class we write. Now we have
the idea of an object that has internal state (like a rational number with a numerator and a denominator). We shall see
that this state is not stored in local variables and does not need to be passed through parameters for methods within the
class. Pay careful attention as we introduce this new location for data and the new ways of interacting with it.
This is quite a shift. Do not take it lightly.
We can create a new object with the new syntax. We can give parameters defining the initial state of the new object.
In our example these are fairly obvious, a numerator and denominator, so we can plan that
Rational r = new Rational(2, 3);

would create a new Rational number for the mathematical expression, 2/3.
Like with built-in types, we can have the natural operations on the type as methods. For instance we can negate
Rational number or convert it to a double approximation, or print it, or do arithmetic.
Thinking ahead to what we would like for our Rational numbers, here is some testing code, with hopefully clear and
reasonable method names:

165
Introduction to Computer Science in C#, Release 2.0

Like other numerical types we would like to be able to parse strings. The helping function, ShowParse, in our testing
code makes the display neater.
One non-obvious method is CompareTo. This one method allows all the usual comparison operators to be used with
the result. We will discuss it more in Rationals Revisited.
The results we would like when running this testing code:
6/(-10) simplifies to -3/5
reciprocal of -3/5 is -5/3
-3/5 negated is 3/5
-3/5 + 1/2 is -1/10
-3/5 - 1/2 is -11/10
-3/5 * 1/2 is -3/10
(-3/5) / (1/2) is -6/5
1/2 > -3/5 ? True
-3/5 as a double is -0.6
1/2 as a decimal is 0.5
Parse "-12/30" to Rational: -2/5
Parse "123" to Rational: 123
Parse "1.125" to Rational: 9/8

One complication with fractions is that there is more than one representation for the same number. As in grade school
we will always reduce to lowest terms.
We are using the same object oriented notation that we have for many other classes: Calls to instance methods are
always attached to a specific object. So far that has always been the object of
object.method( ... )
So far we have been thinking and illustrating how we would like objects in this Rational class to look like and behave
from the outside. We could be describing another library class. Now, for the first time, we start to delve inside, to the
code and concepts needed to make this happen:
Some of the parts are harder than others. We start with the most basic ones. First we need a class.

13.1.2 Class Syntax

Our code is nested inside

public class Rational
{

// ... fields, constructor, code for Rational omitted

This is the same sort of wrapper we have used for our Main programs! Before, everything inside was labeled static.
Now we see what happens with the static keyword omitted....

13.1.3 Instance Variables

A Rational has a numerator and a denominator. We must remember that data. Each individual Rational that we use
will have its own numerator and denominator.
We have used some static variables before in classes, with the keyword static, where there is just one copy for the
whole class. If we omit the static we get an instance variable, that is the particular data for an individual Rational
number. This is our new place to store data:

166 Chapter 13. Classes and Object-Oriented Programming

 Introduction to Computer Science in C#, Release 2.0

We declare these in the class and outside any method declaration.

They are fields of the class. As we will discuss more in terms of safety and security, we add the word “private” at the
beginning:
public class Rational
{
private int num;
private int denom;

// ... constructor, code for Rational omitted

You also see that we are lazy in this example, and abbreviate the long words numerator and denominator in our names.
It is important to distinguish instance variables of a class and local variables. A local variable is only accessible inside
the one method where it was declared, and is destroyed at the end of the method. However the class fields num and
denom are remembered by C# as long as the object is in use. This is a totally new situation. We repeat:

Note: Instance variable have a completely different lifetime and scope from local variables. An object and its instance
variables, persist from the time a new object is created with new for as long as it remains referenced in the program.

We need to get values into our field variables. They describe the state of our Rational.
We have used constructors for built-in types. Now for the first time we create one.

13.1.4 Constructors

The constructor is a slight variation on a regular method: Its name is the same as the kind of object constructed, the
class name (Rational here), and it has no return type (and no static). Implicitly you are creating the kind of object
named, a Rational in this case. The constructor can have parameters like a regular method. We will certainly want to
give a state to our new object. That means giving values to its numerator and denominator. Recall we are want to store
this state in instance variables num and denom. We could just use
public Rational(int numerator, int denominator)
{
num = numerator;
denom = denominator;
}

While the local variables in the formal parameters disappear after the constructor terminates, we want the data to
live on as the state of the object. In order to remember state after the constructor terminates, we must make sure the
information gets into the instance variables for the object. This is the basic operation of most constructors: Copy
desired parameters in to initialize the state in the fields. That is all our simple code above does.
Note that with num and denom we are not using full object notation with an object reference and a dot, as we have
when referring to a field in a different (so far always built-in) class, like arrayObject.Length. The constructor
is creating an object, and the notation for the instance variables is understood to be giving values to the Rational
object being constructed. C# allows this shorthand notation inside a constructor and also inside an instance method
(discussed below).
There is actually a bit more to do in the constructor than we showed: Validate the data. The actual constructor is
The last line calls an instance method, normalize, to make sure the denominator is not 0, and to reduce the fraction
to lowest terms.

13.1. A First Example of Class Instances: Rational 167

Introduction to Computer Science in C#, Release 2.0

Like with the instance variables, with normalize() there is not the full object notation: object.method(). Again C#
is allowing a shorthand: With the explicit object reference missing, the assumption is that the method be applied to the
current object: In this case that is the one just being initialized in this constructor.

13.1.5 Instance Methods

Look at the heading for normalize:

You see that there is no static in the method heading, so the method is attached to the current instance implicitly. It
can only be called when there is a current instance (discussed below). Since the method only deals with the instance
variables, no further parameters need to be given explicitly.
(It is also declared private: It is a helping method only used privately, from inside the class.)
The method uses the GCD Remainder Loop function to reduce the Rational to lowest terms.
The instance variable names and method names used without an object reference and dot refer to the current instance.
Whenever a constructor or non-static method in the class is called, there is always a current object:
1. In a constructor, referring to the object being created.
2. When some instance method methodName is called with explicit dot notation,
someObject.methodName(), then it is acting on the current object someObject.
3. When a constructor or instance method (no static) inside the class is called, there must already be a current
object. If that constructor or instance method calls a further instance method inside the same class, without
using dot notation, then the further method has the same current object.
Again: When a constructor or method refers to or sets an instance variable, without starting with “someObject.”, it is
referring to the current object, also referred to as this object.

Warning: These are the only places where there is a current object. Inside a static method there is no current
object. A common compiler error is to try to have a static method call an instance method without dot notation
for a specific object. The shorthand notation without an explicit object reference and dot cannot be used, because
there is no current object to insert implicitly:

public void AnInstanceMethod()

{
...
}

public static void AStaticMethod() // no current object

{
AnInstanceMethod(); // Compiler error caused
}

On the other hand, there is no issue when an instance method calls a static method. (The instance variables are just
inaccessible inside the static method.)

The current object is implicit inside a constructor or instance method definition, but it can be referred to explicitly. It
is called this. In a constructor or instance method, this is automatically a legal local variable to reference. You
usually do not need to use it explicitly, but you could. For example the current Rational object’s num field could
be referred to as either this.num or the shorter plain num. We will see in later sections that there are places where
there is reason for an object to refer to itself explicitly, and use this as the object’s name.

168 Chapter 13. Classes and Object-Oriented Programming

 Introduction to Computer Science in C#, Release 2.0

13.1.6 Getters

In instance methods, where instance variables are accessible, you have an extra way of getting data in and out of the
method: Reading or setting instance variables. The simplest methods do nothing but that....
We have chosen to make our Rationals immutable. The private in front of the field declarations was important to
keep code outside the class from messing with the values. On the other hand we do want others to be able to inspect
the numerator and denominator, so how do we do that?
Use public methods.
Since the fields are accessible anywhere inside the class’s instance methods, and public methods can be used from
outside the class, we can simply code
These methods allow one-way communication of the numerator and denominator values out of the object. These are
examples of a simple category of methods: A getter simply returns the value of a part of the object’s state, without
changing the object at all.
Note again that there is no static in the method heading. The field value for the current Rational is returned.
Another simple method returns a double approximation:
So far we have returned built-in types. What if we wanted to generate the reciprocal of a Rational? That would be
another Rational. How do we do it? We have a constructor! We can easily use it. The reciprocal swaps the numerator
and denominator:
We have used static methods before (in fact all the methods we wrote before this section were static methods). They
are not associated with an instance. They are still useful. For example, in analogy with the other numeric types we
may want a static Parse method to act on a string parameter and return a new Rational.
The most obvious kind of string to parse would be one like “2/3” or “-10/77”, which we can split at the ‘/’. Integers are
also rational numbers, so we would like to parse “123”. Finally decimal strings can be converted to rational numbers,
so we would like to parse “123.45”. That last case is the trickiest. See how our Parse method below distinguishes
and handles all the cases. It constructs integer strings for both the numerator and denominator, and then parses the
integers. Note that the method is static. There is no Rational being referred to when it starts, but in this case the
method returns one:

13.1.7 Method Parameters of the Same Type

We can deal with the current object without using dot notation. What if we are dealing with more than one Rational,
the current one and another one, like the parameter in Multiply:
We can mix the shorthand notation for the current object’s fields and dot notation for another named object: num and
denom refer to the fields in the current object, and f.num and f.denom refer to fields for the other Rational, the
parameter f.
Note that we do not refer to the fields of f through the public methods GetNumerator and GetDenominator.
Though f is not the same object, it is the same type:

Note: Private members of another object of the same type are accessible from method definitions in the class.

The full Multiply method is:

There are a number of other arithmetic methods in the source code for Rational that return a new Rational result of the
arithmetic operation. They do review your knowledge of arithmetic! They do not add further C# syntax.

13.1. A First Example of Class Instances: Rational 169

Introduction to Computer Science in C#, Release 2.0

13.1.8 ToString Override

All the built-in types can be concatenated into strings with the ‘+’ operators. We would like that behavior with our
custom types, too. How can the compiler know how to handle types that were not invented when the compiler was
written?
The answer is to have common features among all objects. Any object has a ToString method. The default version
supplied by the system is not very useful for an object that it knows nothing about! You need to define your own
version, one that knows how you have defined your type with its own specific instance variables. You need to have
that version used in place of the default: You need to override the default. To emphasize the change in meaning, the
word override must be in the heading:
For any kind of new object that you create and want to be able to implicitly convert to a string, you need a ToString
method with the exact same heading as the ToString for a Rational.
A more complete discussion of override would lead us into class hierarchies and inheritance, which we are not
emphasizing in this book.
The whole code for Rational is in rational_nunit/rational.cs.

13.1.9 Pictorial Playing Computer

Let us start pictorially playing computer on test_rational.cs, as a review of much of the previous sections We
explicitly show a local variable this to identify the current object in an instance method or constructor.
The first line of Main,
Rational f = new Rational(6, -10);

creates a new Rational, so it calls the constructor. At the very beginning of the constructor, a prototype Rational
is already created as the current object, so immediately, there is a this. The parameters 6, and -10 are passed,
initializing the explicit local variables numerator and denominator. The figure illustrates the memory state at
the beginning of the constructor call:

Note the immediate value assigned to the numeric instance variables is zero: This is as discussed in Default Initial-
izations. Of course we do not want to keep those default values: The constructor finds the value of the local variable
numerator, and needs to assign the value 6 to a variable num. The compiler has looked first for a local variable
num, and found none. Then it looked second for an instance variable in the object pointed to by this. It found num
there. Now it copies the 6 into that location. Similarly for denominator and denom:

170 Chapter 13. Classes and Object-Oriented Programming

 Introduction to Computer Science in C#, Release 2.0

Then the constructor calls normalize. Since normalize is also an instance method, a reference to this is passed
implicitly. While illustrating the memory state for more than one active method, we separate each one with a horizontal
segment.

Later normalize calls gcd. Since gcd is static, note that the local variables for gcd do not contain a reference to
this.

At the end of gcd the int 2 is returned and initializes n in the calling method normalize. Then normalize modifies
the instance variable pointed to by this, and finishes.

13.1. A First Example of Class Instances: Rational 171

Introduction to Computer Science in C#, Release 2.0

That is the same object this in the constructor. Just before the constructor completes we have:

Then in Main the constructor’s this is the reference to the new object initializing f.

Consider the next line of Main:

Console.WriteLine("6/(-10) simplifies to {0}", f);

We omit the internals of the WriteLine call, except to note that it must convert the reference f to a string. As with any
object, it does this by calling the ToString method for f, so the implicit this in the call to ToString refers to
the same object as f:

ToString returns “-3/5”, and it gets printed as part of the line generated by WriteLine....
We skip the similar details through two more WriteLine statements and the initialization of h:
Rational h = new Rational(1,2);

The WriteLine statement after that needs to evaluate f.Add(h), generating a call to Add. The next figure shows
the two local variables in Main, f and h, each pointing to a Rational object. The image shows the situation in the
call to Add. In the local variables for the method Add see what the implicit this refers to, and what the (local to
Add) variable f refer to. As the figure shows, this use of a local variable f is independent of the f in Main:

172 Chapter 13. Classes and Object-Oriented Programming

 Introduction to Computer Science in C#, Release 2.0

Since the return statement in Add creates a new object, the figure shows a call to the constructor from inside Add. We
do not go through the details of another constructor call, but this in The constructor ends up pointing to the Rational
shown.
The this of the constructor ends up as the reference returned by Add:

which gets sent to the WriteLine statement and gets printed as in the earlier code.
Make sure you see how the pictures reinforce these important ideas:
• Keeping track of the this with constructors and instance methods (but not static methods).
• The aliasing of Rational objects used as parameters explicitly or implicitly (this).
We have played computer before in procedural programming, following individually explicitly named variables. This

13.1. A First Example of Class Instances: Rational 173

Introduction to Computer Science in C#, Release 2.0

has allowed us to follow loops clearly after the code is written. The pictorial version with multiple object references
and method calls is also useful for checking on code that is written.
When first writing code, a picture of the setup of the references in your data is also helpful. New object-oriented
programmers often have a hard time referring to the data they want to work with. If you have a picture of the data
relationships you can point with a finger to a part that you want to use. For example in the call to Add, one piece of
data you need for your arithmetic is the num field in f. Then you must be carful to note that only local variables can
be referenced directly (including the implicit this). If you want to refer to data that is not a local variable, you must
follow the reference path arrow that leads from a local variable to an instance field that you want to reference.

Then use the proper object-oriented notation to refer to the path. In the example, it takes one step, from local variable
f to its field num, referred to as f.num. Similarly the current object’s num is connected through this, but C#
shorthand allows this. to be omitted. And so on, for f.denom and denom.
Visually following such paths will be even more important later, when we construct more complex types of objects,
and you need to follow a path through several references in sequence.

13.1.10 Local Variables Hiding Instance Variables

A common error is for students to try to declare the instance variables twice, once in the regular instance variable
declarations, and then again in a constructor, like:
public Rational(int numerator, int denominator)
{
int num = numerator; // LOGICAL ERROR!
int denom = denominator; // LOGICAL ERROR!
}

This is deadly. it is worse than redeclaring a local variable, which at least will trigger a compiler error.

Warning: Instance variable only get declared outside of all functions and constructors. Same-name local variable
declarations hide the instance variables, but compile just fine. The local variables disappear after the constructor
ends, leaving the instance variables without your desired initialization. Instead the hidden instance variables just
get the default initialization, 0 for a number.

Generally when you get a compiler error, the error is at or before the location the error is referenced, but with local
variables covering instance variables, the real cause can come later in the text of the method. Below, when you first
refer to r in Badnames, it appears to be correctly referring to the instance variable r:
class ForwardError
{

174 Chapter 13. Classes and Object-Oriented Programming

 Introduction to Computer Science in C#, Release 2.0

private Random r = new Random();

// ...

void BadNames(int a, int b)

{
int n = r.Next(0, 10); // legal in text *just* to here
//...
int r = a % b; // r declaration makes *earlier* line wrong
//...
}

The compiler scans through all of BadNames, and sees the r declared locally in its scope. The error may be marked
on the earlier line, where the compiler then assumes r is the local int variable, which therefore is not an object with
a Next method.
This is based on a real student example. A second issue that it points to is using too short variable names that are not
descriptive of the variable meaning.

13.2 Classes And Structs

C# has an alternate syntax to a class: a struct. Everything we have said so far about classes such an Rational applies
to structs also! In fact you could change class into struct in the heading for Rational, and it would become a
struct, with no further code changes in any of the code we have written!
public struct Rational
{
// ...
}

So why the distinction? We have mentioned that new objects created in a class are accessed indirectly via a reference,
as with an array. As a general category, they are called reference objects. We distinguished the types int and double
and bool, where the actual value of the data is stored in the space for a variable of the type. They are value types. A
struct is also a value type. In practice this is efficient for small objects. We made Rational a class because you have
already seen the class construct with static entries, and classes are more generally useful. In fact being a struct
would be a good choice for Rational, since it only contains two integers. Its size is no more than one double.
The behavior of a Rational is the same either way, because it is immutable. If we allowed mutating methods, then a
class version and a struct version would not behave the same way, due to the fact the reference types can have aliases,
and value types cannot.
There are some more complicated situations where there are further distinctions between classes and structs, but we
shall not concern ourselves with those fine advanced points in this book.

13.3 Class Instance Examples

13.3.1 Getters and Setters

Our class Rational in rational_nunit/rational.cs is a practical utility class. We used it as a first example to avoid being
artificial, and illustrate many points. One choice we made for it to be practical, is to have Rational objects be
immutable: Though we constructed Rationals and used them, there was no public code that modified a Rational. We
discussed Getters. For mutable objects, another basic kind of method is a setter, that sets an attribute of the object’s
state, changing it from what it was before.

13.2. Classes And Structs 175

Introduction to Computer Science in C#, Release 2.0

This time we have a nonsensical but very simple class to illustrate the use of both getters and setters, example exam-
ple_class/example_class.cs. It is highly commented:
Make sure you can follow the code and the output from running.
One important point is disambiguating the use of n and d in several places where they are used as local variable, while
also being instance variables names. For instance note the constructor:
The compiler first looks to see if an identifier, like n is a local variable. If so, it stops there. We need to be trickier to
access an instance variable of the same name. We can generally skip dot notation when referring to the current object,
but here we have an exception, since plain n refers to the local variable. We can use dot notation with a reference to the
current object. The current object is this, so we can refer to this.n. After seeing the this., the compiler knows
that what follows must refer to a part of the current object, and hence the n means the instance variable. Similarly
with d.
In the class Rational we never changed the value of an instance variable in a public method. That kept Rationals
immutable. In Example we choose to have a setter, like SetN, changing the value of the instance variable:
Here, while we are declaring a local variable, the formal parameter, with the same name as the instance variable, we are
careful to use this. to disambiguate the instance variable. The following variation is wrong and is an unfortunately
easy slip:
public void SetN(int newN) // NOT a "setter" method
{
int n = newN; // LOGICAL ERROR! NO LASTING EFFECT
}

This is the same error discussed for a constructor in Local Variables Hiding Instance Variables.
Since an Example object is mutable, we can play with aliases, as in the last few lines of Main, after e becomes an
alias for e2. We change an object under one name, and it affect the alias the same way, as you can see from running
the program.

13.3.2 Converting A Static Game To A Game Instance

For a comparison of procedural and object-oriented coding, consider converting Number Guessing Game Lab so that
a GuessGame is an object, an instance of the GuessGame class.
While our last example, Rational, is in fact a very practical use of object-oriented programming, GuessGame is some-
what more artificial. We create it hoping that highlighting the differences between procedural and object-oriented
presentation is informative. Here is a procedural version, example file static_version/static_version.cs
The project also refers to the library class UI, with the functions we use for safe keyboard input. It is all static methods.
Is there any reason to make this UI class have its own own instances?
No. There is no state to remember between UI method calls. What comes in through the keyboard goes out through
a return value, and then you are completely done with it. A simple static function works fine each time. Do not get
fancy for nothing.
What state would a game hold? We might set it up so the user chooses the size of the range of choices just once, and
remember it for possibly multiple plays of the GuessGame. The variable was big before, we can keep the name. If
we are going to remember it inside our GuessGame instance, then big needs to become an instance variable, and it
will be something we can set in a constructor.
What actions/methods will this object have? Only one - playing a GuessGame. The GuessGame could be played
multiple times, and that action, play, makes sense as a method, Play, which will look a lot like the current static
function.
In the procedural version there are several other important variables:

176 Chapter 13. Classes and Object-Oriented Programming

 Introduction to Computer Science in C#, Release 2.0

• Random rand: That was static before, for good reason: We only need one Random number generator for the
whole time the program is running, so one static variable makes sense.
• The central number in the procedural Game and our future Play method is secret. Should that be an instance
variable? It would work, but it would be unhelpful and misleading: Secret is reset every time the game is played,
and it has no meaning after a Play function would be finished. There is nothing to remember between time you
Play. This is the perfect place for a local variable as we have now.
A common newbie error is to make things into instance variables, just because you can, when an old-fashioned local
variable is all that you need. It is good to have variables leave the programmer’s consciousness when they are no
longer needed, as a local variable does. An instance variable lingers on, leaving extra places to make errors.
This introductory discussion could get you going, making a transformation. Go ahead and make the changes as far as
you can: create project GuessGame inside the current solution. Have a class GuessGame for the GuessGame instance,
with instance variable big and method Play.
You still need a static Main method to first create the GuessGame object. You could prompt the user for the value for
big to send to the constructor. Once you have an object, you can call instance method Play. What about parameters?
What needs to change?
There is also a video for this section that follows all the way through the steps. A possible final result is in in-
stance_version/guess_game.cs.

13.3.3 Animal Class Lab

Objectives: Complete a simple (silly) class, with constructor and methods, including a ToString method, and a
separate testing class.
Make an animal_lab project in your solution, and copy in the files from the example project animal_lab_stub. Then
modify the two files as discussed below.
1. Complete the simple class Animal in your copy of the file animal.cs. The bullets below name and describe the
instance variables, constructor, and methods you need to write:
• An Animal has a name and a gut. In our version the gut is a List of strings describing the contents, in
the order eaten. A newly created Animal gets a name from a parameter passed to the constructor, while
the gut always starts off empty.
• An Animal has a Greet method, so an animal “Froggy” would say (that is, print)
Hello, my name is Froggy.
• An Animal can Eat a string naming the food, adding the food to the gut. If Froggy eats “worm” and
then “fly”, its gut list contains “worm” and “fly”.
• An Animal can Excrete (removing and printing what was first in the gut List). Recall the method
RemoveAt in List Syntax. Print “” if the gut was already empty. Following the Froggy example above,
Froggy could Excrete, and “worm” would be printed. Then its gut would contain only “fly”.
• A ToString method: Remember the override keyword. Make it return a string in the format shown
below for Froggy, including the Animal’s name:
“Animal: Froggy”
• All the methods that print should be void. Which need a parameter, of what type?
2. Complete the file test_animal.cs with its class TestAnimal containing the Main method, testing the class
Animal: Create a couple of Animals and visibly test all the methods, with enough explanation that someone
running the test program, but not looking at the code of either file, can see that everything works.

13.3. Class Instance Examples 177

Introduction to Computer Science in C#, Release 2.0

13.3.4 Planning A Class Structure

We are going to build up to a game for you to write. Here is an idea for a skeleton of a text (adventure?) game. It
does not have much in it yet, but it can be planned in terms of classes. Classes with instances correspond to nouns
you would be using, particularly nouns used in more than one place with different state data being remembered. Verbs
associated with nouns you use tend to be methods. Think how you might break this down. The parts appearing after
the ‘>’ prompt are entries by the user. Other lines are computer responses:
Welcome to Loyola!
This is a pretty boring game, unless you modify it.
Type ’help’ if you need help.

You are outside the main entrance of the university that prepares people for
extraordinary lives. It would help to be prepared now....
Exits: east south west
> help
You are lost. You are alone.
You wander around at the university.

Your command words are:

help go quit

Enter
help command
for help on the command.
> help go
Enter
go direction
to exit the current place in the specified direction.
The direction should be in the list of exits for the current place.
> go west
You are in the campus pub.
Exits: east
> go east
You are outside the main entrance of the university that prepares people for
extraordinary lives. It would help to be prepared now....
Exits: east south west
> go south
You are in a computing lab.
Exits: north east
> go east
You are in the computing admin office.
Exits: west
> bye
I don’t know what you mean...
> quit
Do you really want to quit? yes
Thank you for playing. Good bye.

Think and discuss how to organize things first....

The different parts of a multi-class project interact through their public methods. Remember the two roles of writer and
consumer. The consumer needs good documentation of how to use (not implement) these methods. These methods
that allow the interaction between classes provide the interface between classes. Unfortunately “interface” is used in
more than one way. Here it means publicly specified ways for different parts to interact.
As you think how to break this game into parts, also think how the parts interact.
The code that generated the exchange above is the project folder csproject1.

178 Chapter 13. Classes and Object-Oriented Programming

 Introduction to Computer Science in C#, Release 2.0

The code uses many of the topics discussed so far in this book.
We will add some features from another meaning of Interfaces, and discuss the revision in project csproject_stub (no
1). You might use this as a basis of a project....

13.3. Class Instance Examples 179

Introduction to Computer Science in C#, Release 2.0

180 Chapter 13. Classes and Object-Oriented Programming

 CHAPTER

FOURTEEN

TESTING

Now that we have learned a bit about classes, we’re going to use the same feature to support unit testing. Unit testing
is a concept that will become part of just about everything you do in future programming-focused courses, so we want
to make sure that you understand the idea and begin to make use of it in all of your work.
The notion of unit testing is straightforward in principle. When you write a program in general, the program comprises
what are properly known as units of development. Each language has its own definition of what units are but most
modern programming languages view the class concept as the core unit of testing. Once we have a class, we can test
it and all of the parts associated with it, especially its methods.
We will be introducing parts of file rational_nunit/rational_unit_tests.cs.

14.1 Assertions

A key notion of testing is the ability to make a logical assertion about something that generally must hold true if the
test is to pass.
Assertions are not a standard language feature in C#. Instead, there are a number of classes that provide functions for
assertion handling. In the framework we are using for unit testing (NUnit), a class named Assert supports assertion
testing.
In our tests, we make use of an assertion method, Assert.IsTrue() to determine whether an assertion is success-
ful. If the variable or expression passed to this method is false, the assertion fails.
Here are some examples of assertions:
• Assert.IsTrue(true): The assertion is successful, because the boolean value true is true.
• Assert.IsTrue(false): The assertion is not successful, because the boolean value false is true.
• Assert.IsFalse(false): This assertion is successful, because the test for whether false is equal to false is true.
• Assert.IsTrue(5 > 0): success
• Assert.IsTrue(0 > 5): failure
There are many available assertion methods. In our tests, we use Assert.IsTrue(), which works for everything
we want to test. Other assertion methods do their magic rather similarly, because every assertion method ultimately
must determine whether what is being tested is true or false.

14.2 Attributes

Besides assertions, a building block of testing (in C# and beyond) comes in the form of attributes. Attributes are an
additional piece of information that can be attached to classes, variables, and methods in C#. There are two attributes

181
Introduction to Computer Science in C#, Release 2.0

of interest to us:
• [TestFixture]: This indicates that a class is being used for testing purposes.
• [Test]: This indicates that a method is one of the methods in a class being used for testing purposes.
Without these annotations, classes and methods will not be used for testing purposes. This allows a class to have some
methods that are used for testing while other methods are ignored.
In the remainder of this section, we’re going to take a look at the strategy for testing the Rational class. In general,
your goal is to ensure that the entire class is tested. It is easier said than done. In later courses (Software Engineering)
you would learn about strategies for coverage testing.
Our strategy will be as follows:
• Test the constructor and make sure the representation of the rational number is sound. If the constructor isn’t
initializing an instance properly, it is likely that little else in the class will work properly.
• Then test the rest of the class. Whenever possible, group the tests in some logical way. In the case of the Ra-
tional class, there are three general categories (and one rather special one): arithmetic operations, comparisons,
and conversions. In addition, there is the parsing test, which ensures that we can convert strings representing
fractions into properly initialized (and reduced) rational numbers.
Let’s get started.

14.3 Testing the Constructor

Testing the constructor is fairly straightforward. We essentially test three basic cases:
• Test whether a basic rational number can be constructed. In the above, we test for 3/5, 3/-5, 6/10, and 125. Per
the implementation of the Rational class (how we defined it), these should result in fractions with numerators of
3, -3, 3, and 12; and denominators of 5, 5, 5, and 1, respectively.
• As you can observe from the code, we perform basic assertion testing to ensure that the numerators and denom-
inators are what we expect. For example:
Assert.IsTrue(r.GetNumerator() == 3)

Tests whether the newly minted rational number, Rational(3, 5), actually has the expected numerator of 3.
• If we are able to get through the entire code of the ConstructorTest() method, our constructor test is a
success. Otherwise, it is a failure.
We’ll look at how to actually run our tests in a bit but let’s continue taking a look at how the rest of our testing is done.

14.4 Testing Rational Comparisons

It is pretty well established by now that the ability to compare is of fundamental importance whenever we are talking
about data. Everything we do, especially when it comes to searching (finding a value) and sorting (putting values in
order) depends on comparison.
In this test, we construct a few Rational instances (r1, r2, and r3) and perform at least one test for each of the essential
operators (>, <, and =). Recall from our earlier discussion of the Rational class that we return < 0 when one Rational
is less than another. We return > 0 for greater than, and == 0 for equal to.
If any one of these comparisons fails, this means that we cannot rely on the ability to compare Rational numbers. This
will likely prevent other tests from working, such as the arithmetic tests, which rely on the ability to test whether a
computed result matches an expected result (e.g. 1/4 + 2/4 == 3/4).

182 Chapter 14. Testing

 Introduction to Computer Science in C#, Release 2.0

14.5 Testing Rational Arithmetic

Testing of arithmetic is a fairly straightforward idea. For all of these tests, we create a couple of rational numbers (47/64
and -11/64) and then call the various methods to perform addition, subtraction, multiplication, division, reciprocal, and
negation.
The key to testing arithmetic successfully in the case of a Rational number is to know know what the result should be.
As a concrete example, the result of adding these two rational numbers should be 36/64. So the testing strategy is to
use the Add() method to add the two rational numbers and then test whether the result of the addition is equal to the
known answer of 36/64.
As you can observe by looking at the code, the magic occurs by checking whether the computed result matches the
constructed result:
Assert.IsTrue(r.CompareTo(new Rational(36, 64)) == 0);

Because we have separately tested the constructor and comparison methods, we can assume that it is ok to rely upon
comparison methods as part of this arithmetic test.
And it is in this example where we begin to see the art of testing. You can write tests that assume that other tests of
features you are using have already passed. In the event that your assumption is wrong, you’d be able to know that
this is the case, because all of the tests you assumed to pass would not have passed.
Again, to be clear, the arithmetic tests we have done here assume that we can rely on the constructor and the comparison
operation to determine equality of two rational numbers. It is entirely possible that this is not true, so we’ll be able
to determine this when examining the test output (we’d see that not only the arithmetic test fails but possibly the
constructor and/or comparison tests as well).
The remaining tests are fairly straightforward. We’ll more or less present them as is with minimal explanation as they
are in many ways variations on the theme.

14.6 Testing Rational Conversions (to other types)

In this test, we want to make sure that Rational objects can be converted to floating point and decimal types (the built-in
types of the C# language).
For example, Rational(3/6) is 1/2, which is 0.5 (both in its floating-point and decimal representations.

14.7 Testing the Parsing Feature

The parsing test tests whether we can convert the string representation of a rational number into an actual (reduced)
rational number. We test three general cases:
• The ability to take a fraction and convert it into a rational number. This fraction may or may not have a “-” sign
in it. For example -12/30 should be equivalent to constructing a Rational(-12, 30).
• The ability to take a whole number and get a proper Rational, e.g. 123 is equal to Rational(123)
• The ability to take a textual representation (1.125) and get a proper Rational(9, 8) representation. In this case,
we are also getting an extra test to ensure the result is reduced.

14.5. Testing Rational Arithmetic 183

Introduction to Computer Science in C#, Release 2.0

14.8 Running the Tests

In Xamarin Studio, run the Run rational_nunit project. A test pad should appear and show something like

As you can see in the above display, all of the tests in RationalTests get executed, and they all pass.
We waited until now to discuss unit testing, because the test classes are coded with instance methods, unlike the static
methods that we started out with.

184 Chapter 14. Testing

 CHAPTER

FIFTEEN

INTERFACES

15.1 Rationals Revisited

C# has a built-in method to sort a List. List is a generic type, however, so how does C# know how to do comparisons
for all different types? Is this specially programmed in for built-in types, or can it be extended to user-defined types?
In fact it can be extended to user defined types, such as our Rational. To sort objects, you only need to be able to do
one thing: indicate which object comes before another. We can do that. The CompareTo method already does that.
If Rational r1 is less than Rational r2, then
r1.CompareTo(r2) < 0

The single CompareTo method is very versatile: Just by varying the comparison with 0, you vary the corresponding
comparison of Rationals:
r1.CompareTo(r2) < 0 means r1 < r2
r1.CompareTo(r2) <= 0 means r1 <= r2
r1.CompareTo(r2) > 0 means r1 > r2
r1.CompareTo(r2) >= 0 means r1 >= r2
r1.CompareTo(r2) == 0 means r1 is equal to r2
r1.CompareTo(r2) != 0 means r1 is not equal to r2
None of the other methods for Rationals make any difference for sorting: Just this one method is needed. Of course
the comparison of strings or doubles are done with totally different implementations, but they have methods with the
same name, CompareTo, and with the same abstract meaning. Still C# is strongly typed and we are talking about
totally different types.
An interface allows us to group diverse classes under one interface type. An interface just focuses on the commonality
of behavior in one or more methods among the different classes. In this case we are only concerned with one method,
CompareTo. We want it to be able to compare to another object of the same type.
C# defines an interface IComparable<T>. A type T can satisfy this interface if if has a public instance method with
signature:
public int CompareTo(T other);

There is one more step before we can use a library method to sort: Although this is the name that C# requires to be
able to satisfy the Icomparable<T> interface, it does not automatically assume that is your intention. You must
explicitly say you want your class to be considered to satisfy this interface. For instance for Rational, we need to
change the class heading to:
public class Rational : IComparable<Rational>

185
Introduction to Computer Science in C#, Release 2.0

In general one or more interface names can be listed after the class name and a colon, and before the opening brace
of the class body. This particular interface is defined in System.Collections.Generic, so we need to be using that
namespace.
The project interfaces has the modified rational.cs and test_rational_sort.cs to test this with a list of Rationals:
which prints:
Before sorting: 1/2 11/3 -1/10 2/5 2/3 1/3
After sorting: -1/10 1/3 2/5 1/2 2/3 11/3

15.2 csproject Revisited

The csproject1 project skeleton was set up with the different commands in different classes, keeping related things
together.
On the other hand they had high level structure in common. Similar names were consciously used for methods:
• Each command needed to Execute
• Each command needed to have Help for the user.
The corresponding names made it somewhat easier to follow the part of the Game constructor with the additions to the
helpDetails Dictionary. Also there is repetitive logic in the crucial proccessCommand method.
In a game with more possible commands, the code would only get more repetitious! You would like to think of having
a loop to go through repetitious code.
A major use of a C# interface will allow this all to work in neat loops. For the first time we define our own interface,
and use that interface as a type in a declaration.
While we are at this we can refactor our code further: classes that give a response to a command all obviously have
their Execute and Help methods. They also have a command word to call them. We can further encapsulate all data
for the response by having the classes themselves be able to announce the command that calls them. We add a string
property CommandWord to each of them.
We will add an extra convenience feature of C# here. Thus far we have used private instance variables and public
getter methods. We can use a public instance variable declaration with a similar effect as in:
public string CommandName {get; private set;}

The extra syntax in braces says that users in a another class may freely get (read) the variable, but setting the variable
is still private: it may only be done inside the class. This is more concise than using a getter method: No getter needs
to be declared; referencing the data is shorter too, since it is a property, no method parentheses are needed.
Note the unusual syntax: the declaration does not end with a semicolon. The only semicolons are inside the braces.
You will not be required to code with this notation, but it sure is neater than using a getter method!
Now we can define our own interface taking all of these common features together. Since each is a response to a
command, we will call our interface Response:
Things to note:
• The heading has the reserved word interface instead of class.
• All the common method headings and the property declaration are listed.
• See what is missing! In place of each method body is just a semicolon.
• Everything in an interface is public. The part of the property about private access is merely omitted.

186 Chapter 15. Interfaces

 Introduction to Computer Science in C#, Release 2.0

We are going to need a collection if we want to simplify the code with loops. We could use code like the following,
assuming we already declared the objects helper, goer, and quitter:
Response[] resp = {helper, goer, quitter};

See how we use Response as a declaration type! Each of the objects in the declaration list is in fact a Response.
Now that we can process with this collection and foreach loops, we do not need the object names we gave at all:
We can just put new objects in the initialization sequence!
Now that we can think of these different objects as being of the same type, we can see the processCommand logic, with
its repetitive if statement syntax is just trying to match a command word with the proper Response, so a Dictionary
is what makes sense!
In fact all the logic for combining the various Responses is now moved into CommandMapper, and the the Com-
mandMapper constructor creates the Dictionary used to look up the Response that goes with each command word.
Here is the whole code for ResponseMapper, taking advantage of the Dictionary in other methods, too.
There is even more to recommend this setup: The old setup had references in multiple places to various details about
the collection of Responses. That made it harder to follow and definitely harder to update if you want to add an new
command. Now after writing the new class to respond to a new command, the only thing you need to do is add a new
instance of that class to the array initializer in the CommandMapper constructor!
The revised Xamarin Studio project is csproject_stub (no 1 this time).
See how the Game class is simplified, too.
Talking about adding commands - these classes could be the basis of a game project for a small group. Have any
ideas?

15.2.1 Cohesion, Coupling, and Separation of Concerns

There are three important ideas in organizing your code into classes and methods:
Cohesion of code is how focused a portion of code is on a unified purpose. This applies to both individual methods
and to a class. The higher the cohesion, the easier it is for you to understand a method or a class. Also a cohesive
method is easy to reuse in combination with other cohesive methods. A method that does several things is not useful
to you if you only want to do one of the things later.
Separation of concerns allows most things related to a class to take place in the class where they are easy to track,
separated from other classes. Data most used by one class should probably reside in that class. Cohesion is related to
separation of concerns: The data and methods most used by one class should probably reside in that class, so you do
not need to go looking elsewhere for important parts.
Some methods are totally related to the connection between classes, and there may not be a clear candidate for a class
to maximize the separation of concerns. One thing to look at is the number of references to different classes. It is
likely that the most referred to class is the one where the method should reside.
Coupling is the connections between classes. If there were no connections to a class, no public interface, it would
be useless except all by itself. The must be some coupling between classes, where one class uses another, but with
increased cohesion and strong separation of concerns you are likely to be able to have looser coupling. Limiting
coupling makes it easier to follow your code. There is less jumping around.
Aim for strong cohesion, clear separation of concerns, and loose coupling. Together they make your code clearer,
easier to modify, and easier to debug.

15.2. csproject Revisited 187

Introduction to Computer Science in C#, Release 2.0

IGame Interface Exercise

On a much smaller scale than the project, this exercise offers you experience writing classes implementing and using
an interface
1. Copy project stub igame_stub to your own project, and note the additions discussed below.
2. Look at the IGame interface in i_game.cs. Then look at addition_game.cs, that implements the interface. See
how a new AdditionGame can be added to list of IGame‘s. Run play_games.cs. Randomly choosing
a game when there is only one to choose from is pretty silly, but it gives you a start on a more elaborate list
of games. The PopRandom method is a good general model for choosing, removing, and returning a random
element.
3. Write several very simple classes implementing the IGame interface, and modify Main in play_games.cs to
create and add a new game of each type. (Test adding one at a time.)
One such game to create with little more work would be a variation on instance based Guessing Game in-
stance_version/guess_game.cs. You need to make slight modifications. You could make Play return the opposite
of the number of guesses, so more guesses does generate a worse score.

15.3 Group Game Project

15.3.1 Objectives

• Being creative, imagining and describing a program, and working it through to completion
• Working in a team:
– Communicating to each other
– Dividing responsibilities
– Helping each other
– Finding consensus
– Dealing with conflict
– Providing process feedback
– Integrating parts created by several people
• Developing new classes to fit your needs
• Using the .Net API library
• Designing a program for refinement
• Testing
• Evolving a program
• Creating documentation for user and implementers
• Programming in the large – not a small predefined problem
• Make something that is fun
See the project csproject_stub, already discussed in class.
What to turn in:
• Periodic reports

188 Chapter 15. Interfaces

 Introduction to Computer Science in C#, Release 2.0

• intermediate implementation
• a final presentation
• a final implementation including user and programmer documentation and process documentation
• individual evaluations

15.3.2 Overview

You will be assigned to groups of 3-4 people to work on a project that will extend until the end of the semester, with
only a little other work introduced in class, and of course Exam 3. This leaves a month of course time for the project
(classes, labs, dedicated homework time), ending with presentations in final exam period. Your group will be designing
and implementing a game. You are invited to start from the project that is provided and has been discussed in class, or
start from scratch if you really want.
Develop a text based game driven by user commands, involving moving between different locations or game states.
The final game should be fun to play and have some goal – some way to win it. The game may be a game for one
person working against what is built into the program, or it may involve several players.
Start by brainstorming and listing ideas – do not criticize ideas at this point. That is what is meant by brainstorming -
having your internal critic going inhibits creativity. After you have a large list from brainstorming, it is time to think
more practically and settle on one basic situation, and think of a considerable list of features you would like it to
have. Order the features, considering importance, apparent ease of development, and what depends on what else. Get
something simple working, and then add a few features at a time, testing the pieces added and the whole project so
far. Test, debug, and make sure the program works completely before using your past experience to decide what to
add next. This may different than what you imagined before the work on your first stage! Like the provided project,
early stages do not need to be full featured, but make sure that you are building up to a version that you can win, and
which includes interesting features. You should end with a game that has enough instructions provided for the user,
so someone who knows nothing of your implementation process or intentions can play the game successfully. Your
implementation should also be documented, imagining that a new team of programmers is about to take over after
your departure, looking to create yet another version.

15.3.3 Your Team

Your instructor will tell you about team makeup.

There are a number of roles that must be filled by team members. Some will be shared between all members, like
coder, but for each role there should be a lead person who makes sure all the contributions come together. Each person
will have more than one role. All members are expected to pull their own weight, though not all in exactly the same
roles. Everyone should make sustained contributions, every week, documented in the weekly reports. Understand that
this project will be the major course commitment for the rest of the semester. These roles vary from rather small to
central. Not all are important immediately.

15.3.4 Roles

1. Leader: Makes sure the team is coordinated, encourage consensus on the overall plan, oversee that the agree-
ments are carried through, be available as contact person for the team and the TA and your instructor.
2. Lead programmer: Keep track of different people’s parts to make sure they fit together.
3. Coder/unit tester: Everyone must have a significant but not necessarily equal part in this job. Each person should
have primary responsibility for one or more cohesive substantive units, and code and test and be particularly
familiar with those parts. Do your best to make individual parts be cohesive and loosely coupled with other
people’s work, to save a lot of pain in the testing and debugging phase. When coding you are still encouraged

15.3. Group Game Project 189

Introduction to Computer Science in C#, Release 2.0

to do pair programming, though what pair from the team is working together at different times may be fluid.
The lead programmer might be involved in pairs more often than others, but be sure the other coders get to drive
often.
4. Librarian/version coordinator: The default should be for you to have a box.com folder shared with your whole
team and me and your TA, with all as editors. Your folder should have the name of your team. You should
always have a folder that contains the latest working version of the project. You should also keep old versions,
for instance copied into numbered version folders. Box does not handle successive versions automatically. You
can choose to use the more capable professional combination of Bitbucket and Mercurial and Teamwork. The
latter will have a learning curve, and in that case this person should be the best informed on Mercurial, and help
the other team members.
5. Report coordinator: Gather the contributions for reports from team members and make sure the whole reports
get to posted on schedule. Your instructor needs a clear idea of the contributions of each member each week. If
a team member is not clear on this to the report writer, the report writer needs to be insistent.
6. Instruction coordinator: Make sure there are clear written documents and help within the program for the user,
who you assume is not a C# programmer and knows nothing about your program at the start.
7. Documentation coordinator: Make sure the documentation is clear for method users/consumers. This includes
the documentation for programmers before the headings of methods and classes. This is for any time someone
wants to use (not rewrite) a class or method you wrote. Also have implementer documentation, for someone who
will want to modify or debug your code and needs to understand the details of your internal implementations.
The extent of this can be greatly minimized by good naming.
8. Quality manager: Take charge of integrated tests of the whole program (following coder’s unit tests). Make sure
deficiencies are addressed.
Conflict resolution: You will certainly have disagreements and possibly conflicts. Try to resolve them within the team.
When that is not working, anyone can go to the instructor with a problem.

15.3.5 The process

Initial:
1. Agree on roles. These roles can change if necessary, but you are encouraged to stick with them for simplicity
and consistency.
2. Agree on a team name and a short no-space abbreviation if necessary, and let me know it.
3. Brainstorm about the project. Distill the ideas into a direction and overall goals.
On individual versions (Two formal versions will be required):
1. Break out specific goals for the version. How are you heading for your overall goals? Are you biting off a
significant and manageable amount? You are expected to check in with me on this part and 2 and 3 before
moving very far. This will be new for most of you.
2. Plan and organize the necessary parts and the interfaces between the parts.
3. Write the interface documentation for consumers of the code for the parts you plan to write. Agree on them.
You need to do this eventually anyway. Agreement up front can save you an enormous amount of time! Do not
let the gung-ho hackers take off before you agree on documented interfaces. We have seen it happen: If you do
not put your foot down, you are stuck with a bad plan that will complicate things. Otherwise lots of code needs
to be rethought and rewritten.
4. If more than one person is working on the same class, plan the names, meanings, and restrictions on the private
instance variables – all coders should be assuming the same things about the instance variables! Also agree on
documentation for any private helping methods you share.
5. Code to match the agreed consumer interface and class implementation designs.

190 Chapter 15. Interfaces

 Introduction to Computer Science in C#, Release 2.0

6. Check each other’s code.

7. Do unit tests on your own work, and fix them and test again...
8. Do overall tests of everything together, and fix and test again...
9. Look back at what you did, how it went, what you could do better, and what to change in your process for the
next version.
You are strongly encouraged to follow modern programming practice which involves splitting each of these formal
versions into much smaller steps that can be completed and tested following a similar process. Order pieces so you
only need to code a little bit more before testing and combining with pieces that already work. This is enormously
helpful in isolating bugs! This is really important. If you thought you spent a long while fighting bugs in your small
homework assignment, that is nothing compared to the time you can spend with a large project, particularly if you
make a lot of haphazard changes all at once.

15.3.6 Splitting Up The Coding

Make good use of the separation of public interface and detailed implementation. If your project has loosely coupled
class, the main part of the public interface should be limited and easy to comprehend.
Ideally have one individual (or pair) assigned a whole class. One useful feature for allowing compiling is to first
generate a stub file like we have given you for homework, that includes the public interface documentation, headings,
and dummy return values and compiles but does nothing. Post this under a box folder for the current version number.
You will then provide your team members with something that tells them what they can use and allows them to compile
their own part. Then later substitute more functional classes.
Your instructor and you will want to review your code. We do not want to have to reread almost the same thing over
and over: Use the editor copy command with extreme caution. If you are considering making an exact copy, clearly
use a common method instead. If you copy and then make substitutions in the same places, you are likely better off
with a method with the common parts and with parameters inserted where there are differences. You can make a quick
test with a couple of copied portions, but then convert to using a method with parameters for the substitutions. Besides
being a waste of effort to define seven methods each defining a tool, with just a few strings differing from one method
to the next, we will require you to rewrite it, with one method with parameters, and just seven different calls to the
method with different parameters. Save yourself trouble and do it that way the first time, or at least after you code a
second method and see how much it is like the first one you coded....
If you are making many substitutions of static textual data, put the data into a resource file in a variation of the Fake
Advise Lab.
You only want to commit working code into the shared current version folder. Comment out incomplete additions that
you want to show to everyone, or comment out the call to a method that compiles but does not yet function logically.
An alternative is to have a separate folder for in-process code to share for comment, so you will not try to compile it
with the current working version.

15.3.7 Weekly reports

Reports are due from the report writer each Tuesday.

1. Inside your team’s box folder have a subfolder called WeeklyReports. A sample stub form to fill out on the
computer is in Weekly-Report.rtf. Make the name of each weekly report document be the date it was due, like
Mar26.rtf. It is easy to copy the table from this week to last week and edit it to show how much your plans
matched reality. You should post a version for your team to look at first. Please distinguish drafts from the final
version for me to look at. You might have a separate folder Drafts, and move the report into the WeeklyReport
folder when it is final. Box easily allows moving files, but not renaming them.

15.3. Group Game Project 191

Introduction to Computer Science in C#, Release 2.0

2. Only one report should be generated each week, with the person in the role of report writer making sure a
complete version is produced and placed in the WeeklyReport folder.
3. Under plans for the next week, include concrete tasks planned to be completed, and who will do them, with
an informative explanation. The content and depth of the person’s work should be clear. If you can state that
clearly and be brief, great. The tasks do not only include coding: they can be any of the parts listed above, and
for any particular part of the project, where that makes sense. If individuals cannot state clearly what they are
working on, then the team leader and lead coder have a significant issue in their leadership that needs to be
addressed.
4. In the review of the last week (after the first week) include the last week’s plans and what actually happened,
task by task, concretely, with enough detail to give an idea on the magnitude of the work. This can include the
portion completed and/or changes in the plans and their reasons. “Still working on X” is not useful: Who was
doing what? What methods, doing what, were completed? Which are in process? Which are being debugged?
What part remains to be done, and who is it assigned to? The report writer is responsible to get a clear statement
from each team member.

15.3.8 Intermediate deliverables

These materials should be placed in a subfolder Intermediate of your team project folder. See the due date in the
schedule.
• Include parts 2-4 listed below under Final Deliverables, but for an intermediate version that runs, and does not
need to have the goal working yet. Have documentation of your methods, including summary description and
description of parameters and return values. If for some reason you do not have all the documentation that
you were encouraged to write first, at least be sure to have and point out significant examples of your clear
documentation of purpose, parameters, and return values. This allows instructor feedback for completing the
rest.
• Copy the linked stub of projectPlans.rtf document. Then complete it:
– List the project roles again, and who ended up filling them. For coding, say who was the person primarily
responsible for each part.
– If you used old classes, like those from the skeleton project or a lab or somewhere else, say which ones are
included unchanged or give a summary of changes.
– If your documentation of methods is not generally done, say what classes got clear documentation (or
individual methods if only some were done).
– Where are you planning to go from here, and who you envision being primarily responsible for different
parts?
• The idea is to have everyone get an idea of what is expected, so we have no misunderstandings about the final
version. We will give you feedback from this version to incorporate in the final version. We do not want to have
to say anyone did anything “wrong” on the final version. We want to be able to concentrate on your creative
accomplishments.
• Look through the list of deliverables again and make sure your collection is complete.

15.3.9 Final Deliverables

Group Submission:
One submission of the group work is due one hour before the final presentations.

192 Chapter 15. Interfaces

 Introduction to Computer Science in C#, Release 2.0

1. All files listed in parts 2-5. Also include a zip file, named with your team abbreviation, containing a Windows
executable with (a separate copy of) any other image and data files needed. Test to make sure you can unzip and
run the executable. The final submissions will be accessible to the whole class – so we can all play them!
2. Source code. You can name the classes appropriately for the content of your game.
3. User instructions. These should be partly built into the program. The most extensive documentation may be
in a document file separate from the program, if you like. (Plain text, MS Word, Rich text (rtf), or PDF,
please.) The starting message built into the beginning of the game should mention the file name of such external
documentation, if you have it.
4. Programmer documentation. Document the public interface for all methods in comments directly before the
method heading. Add implementation comments embedded in the code where they add clarity (not just ver-
bosity). You may have a separate overview document. Include “Overview” in the file name
5. Overall project and process review in a document named like the linked stub, projectReview.rtf.
• The first section should be Changes. So the instructor does not duplicate effort, please give an overview
of the changes from the intermediate version. What classes are the same? What features were added?
What classes are new? Which classes or methods were given major rewrites? What classes had only a few
changes? (In this case try to list what to look for.)
• List again the roles, and who filled them. For coding, say who was the person primarily responsible for
each part.
• What did you learn? What were the biggest challenges? What would you do differently the next time?
What are you most proud of?
• How could we administer this project better? What particularly worked about the structure we set up?
6. A 10-15 minute presentation of your work to the class in final exam period. What would you want to hear about
other projects? (Say it about yours.) What was the overall idea? What was the overall organization? What did
you learn that was beyond the regular class topics that others might find useful to know? What were your biggest
challenges? Do not show off all your code just because it is there. Show specific bits that gave you trouble or
otherwise are instructive, if you like.
Look through the list of deliverables again, before sending files, and check with the whole team to make sure your
collection is complete.
Your Assessment of Individuals in the Group:
This is due electronically 10 minutes after the final class presentation period, from each team member, independently,
turned in a manner specified by your instructor, like other homework assignments.
Change the name of the linked stub file Indiv-Mem-Assessment.rtf to your teamAbbreviation-yourName.rtf. You may
want to tweak it after the group presentation, but have it essentially done beforehand.
Writing this is NOT a part of your collective group deliberations. It is individual in two senses: both in being about
individual team members and in being the view of one individual, you. For this document only, everyone should be
writing separately, privately, and independently from individual experience. If you lack data on some point, say so,
rather than using what others are saying.

15.3. Group Game Project 193

Introduction to Computer Science in C#, Release 2.0

194 Chapter 15. Interfaces

 CHAPTER

SIXTEEN

RECURSION

We are looking forward to a data structures course. Recursion is an important topic. It is not a topic we require in the
introductory course.
More later.

195
Introduction to Computer Science in C#, Release 2.0

196 Chapter 16. Recursion

 CHAPTER

SEVENTEEN

DATA STRUCTURES

We have discussed some basic data structures that are available in C#: Array, List, Set, Dictionary. There are many
more with many implementations. Further study leads you into a data structures course.
We may add some optional, forward looking material here.

197
Introduction to Computer Science in C#, Release 2.0

198 Chapter 17. Data Structures

 CHAPTER

EIGHTEEN

APPENDIX

18.1 Development Tools

18.1.1 About Software Development Kits (SDKs)

A software development kit (SDK) is a set of tools for developing in a particular programming language (in our class,
C#). Developing in a language means everything from compiling to running and (when things go wrong) to debugging
programs.
The Microsoft SDK is the proprietary implementation of .Net. It runs only on Windows and is the primary development
framework for all things Microsoft.
The Mono Project SDK <http://mono-project.com> is the free/open source equivalent implementation of the Microsoft
SDK. It runs on all major platforms (including Windows) and is needed in situations where you want to develop .Net
applications on non-Windows platforms.
As an interesting aside, the company whose developers lead the work on the Mono SDK are working on commercial
tools that allow you to develop/run applications written in .Net on Apple iOS and Android mobile devices (phones and
tablets).

18.1.2 Editing and Building Tools

Early programs were written with rudimentary text editors, more primitive than Windows Notepad. Gradually tools
got better. Now there are editors that are highly optimized for editing code.
After code is edited, it has to be converted into an executable program. That may involve several files and libraries
and other dependencies. Streamlining and automating this process was a big deal. There are a variety of building tools
that can be used with, or built into an SDK: make, ant, and now NAnt for .net.
Many developers use an a la carte approach, using their favorite editor along with their favorite building tool.

18.1.3 About Integrated Development Environments (IDE)

There are also all-in-one tools that combine an editor and build tools. These are also used by many developers.
There are two major IDEs for .Net development, which we explain briefly below:
• Visual Studio is the Microsoft IDE that interfaces directly to the Microsoft SDK.
• Xamarin Studio is the free/open source IDE for developing applications using the Mono SDK on Windows and
all other platforms (in particular, Linux and OS X). The project started as MonoDevelop. Now Xamarin is both
a major contributor to the code and has commercial versions for iOS development. The name on the software is
now Xamarin Studio, though you may see references to MonoDevelop instead.

199
Introduction to Computer Science in C#, Release 2.0

In addition, there is another Windows-specific IDE, SharpDevelop, that inspired the creation of Xamarin Studio. It
is still actively maintained and provides a somewhat “lighter weight” alternative to Visual Studio for Windows users.
Like Xamarin Studio, it is aimed at developers who would prefer a more free/open source “friendly” version.

18.1.4 Our Approach

In the interest of providing a consistent experience for our students who use various operating systems on their on=wn
machines, we will be using the multi-platform Mono (the SDK).
We find the IDE Xamarin Studio convenient to integrate everything for a beginner, and it is a powerful tool at a more
advanced level. Hence we start off introducing and using Xamarin Studio. Later we will look at some of the underlying
tools that are obscured by the use of Xamarin Studio.
Mono has an extra advantage in the tool csharp, for immediate testing of small snippets of code. We will ude it
extensively as we introduce bits of syntax.
As there is significant evolution of both the Microsoft and Mono toolchains–a fancy word we want you to know and a
more elegant way of saying SDK–we’ll issue updates to this book.

18.1.5 Installing Mono

Because the Mono Project web page is known to change frequently, these instructions are designed to be as generic as
possible. If you have any questions, you should contact the instructors immediately or seek tutoring help.

18.1.6 OS X

1. Go to <http://mono-project.com>.
2. Look for the Mono downloads link. You want to get the latest stable version of Mono for OS X. For this class,
you need version 2.10 or later.
3. You may see a link to download Runtime or SDK. Make sure you select SDK.
4. For OS X, the SDK is distributed as a DMG disk image. You’ll need to download this image and double-click
it. Open the image and run the installer. Administrative privileges are required to run the installer, so if you do
not know this information, please stop here.
Here is how to do a quick sanity check of your Mono setup:
1. Go to Applications -> Utilities and launch the Terminal application. (Terminal is how you get to a command-line
shell in OS X.)
2. You’ll see a prompt that looks like this computername:folder user$. This means that Terminal is ready
for input.
3. Type which csharp and hit enter/return. You should see /usr/bin/csharp as output. csharp is the
C# interpreter.
4. Type which mcs and hit enter/return. You should see /usr/bin/mcs as output. mcs is one of the interfaces
to the C# compiler.

Xamarin Studio Installation

1. Make sure Mono is installed first.

2. Now go to <http://monodevelop.com> (not Xamararin for the open source version).

200 Chapter 18. Appendix

 Introduction to Computer Science in C#, Release 2.0

3. As with Mono, we need to look for the downloads link. You should download the stable version.
4. As with Mono, you will see a DMG file, which you should download and double-click to mount on your desktop.
5. This time, you will see an App for Xamarin Studio, which you can drag and drop into the Applications folder.
6. If the preceding steps were successful, you can launch Xamarin Studio by double-clicking the icon in your
Applications folder. (You won’t know what to do with it yet, but at least you can verify that it launches correctly
and then use Command-Q to exit.)

18.1.7 Windows

1. Go to <http://mono-project.com>.
2. Look for the Mono downloads link. You want to get the latest stable version of Mono for Windows. For this
class, you need version 2.10 or later.
3. You may see a link to download Runtime or SDK. Make sure you select SDK.
4. For Windows, there is only one option to download the SDK. It is a self-extracting executable, so you will need
to double click it to install. For Windows 7 users, you may need to check your taskbar to see whether the installer
is being held up by Microsoft’s enhanced security, UAM, that makes sure you really want to install something
you downloaded from the internet.
Here is how to do a quick sanity check of your Mono setup:

Mono Command Prompt

1. Open the Windows Start Menu and type “mono” in the text field at the bottom. You should see a short list of
places “mono” appears.
2. Click on the choice that says “Mono ... Command prompt”. (This is probably faster than going to the Start
Menu, finding the Mono folder, expanding it, and clicking on the Mono Command Prompt.)
If it comes up, you are all set for an initial installation check. This will be the first step later, when you want to run the
handy csharp program or compile and run your own programs. When working, you can just leave this window open,
saving it for later use, (or close and reopen later....)

Xamarin Studio Installation

1. Have Mono installed first.

2. Now go to <http://mono-develop.com>.
3. As with Mono, we need to look for the downloads link. You should download the stable version. That should
be at least numbered 4.0.8.
Install the second and third packages first, then the Xamarin Studio installer.
4. As with Mono, you will see a self-extracting installer, which you should run as before.
5. If the preceding steps were successful, you can launch Xamarin Studio by double-clicking the icon in your
Applications folder. (You won’t know what to do with it yet, but at least you can verify that it launches correctly
and then close the window.)

18.1. Development Tools 201

Introduction to Computer Science in C#, Release 2.0

18.1.8 Linux

We only provide instructions for Debian-based Linux distributions such as Ubuntu.

1. Using the command-line apt-get tool, you can install everything that you need using apt-get install
monodevelop. This should be run as the root user (using the sudo command).
2. You can test the sanity of your setup by following the instructions under OS X.
Xamarin Studio releases on Linux tend to lag behind the official stable release.
This page, https://launchpad.net/~keks9n/+archive/monodevelop-latest,
describes how to update your Xamarin Studio setup if it is not version 2.8 or later as we’ll need for this course.
We wish to stress that Linux is recommended for students who already have a bit of programming experience under
their belts. It can take a significant amount of energy to get a Linux setup up and running and to tweak it to your liking.
While it has gotten ever so much easier since the 1990s when it first appeared, we encourage you to set it up perhaps a
bit later in the semester or consider running it using virtualization software (on Mac or Windows) such as VirtualBox
or VMware.

18.2 Xamarin Studio

Several sections have given basics for Xamarin Studio:

Xamarin Studio Installation
Lab: Editing, Compiling, and Running with Xamarin Studio
:ref: xamarinstudio-reminders
Running our Xamarin Studio Examples Solution

18.3 Command Line Introduction

1
We will be directing you to use a command window or terminal to compile and run C# programs.
Reasons to use the command line:
• The command line precedes the graphical user interface (GUI) used in modern operating systems and provides
a simpler interface for input and output that is very flexible and powerful for knowledgeable users.
– Input comes from the keyboard as typed characters (no mouse processing). Commands are only processed
once you press Enter.
– Output goes to the monitor as textual information (no window processing).
– In C# these input/output mechanisms are called Console processing.
– Input from and output to files is done in a very similar way, simplifying learning.
• Many software development organizations use command line processing to automate creating, compiling
(“building”), and running or executing software programs.
– Command line “scripts” can be created to automate routine tasks.
– Command line scripts are similar to C# and other computer programs.
– Serious software developers should be familiar with the command line.
1 Thanks to Dr. Robert Yacobellis for elaborations to this section.

202 Chapter 18. Appendix

 Introduction to Computer Science in C#, Release 2.0

Accessing the command line (often called a command shell):

• On Windows, to have easy access to Mono tools, press the Windows key (lower left or lower right on the
keyboard) and type Mono, then select the Mono Command Prompt and press Enter.
Alternately, if you do not need Mono tools for sure, the general way to get a command window is to press the
Windows key and R (lower or upper case) together, then type cmd and press the Enter or Return key – this brings
up the basic command processing program, cmd.
• On a Mac just open a Terminal window – this is fine for the Mono SDK commands.
Mac OSX, Linux and other Unix variants work basically the same once you get to a terminal, so we will only distin-
guish Windows and Mac OS-X.

18.3.1 Navigating Directories

First make sure you are familiar with Path Strings.

In a command shell there is always a current working directory, usually shown in the prompt for the next command.
When you open a Mono Command Prompt or Terminal window you will see a prompt that tells you what folder or
directory the command shell has started in: In Windows 7 this is typically C:\Windows\System32, and on a Mac it is
typically /Users/yourLogin.
Files in the current working directory can to referred to by their simple names, e.g., myfile.txt. You can list all the files
in the directory with the simple command dir (short for directory) in Windows or ls (short for list) on a Mac.
You will see below that you can change the current directory with the cd command.
You need to refer to files not in the current directory via a relative or absolute path name.
On a Mac, the file system is unified in one hierarchy. On Windows there may be several drives, and you need to start
a path reference with a drive, like C:, if it is not the current drive.
When you open a new command window in Windows, the starting directory is unfortunate,
C:\windows\System32. You are likely to want to operate out of your home directory (where the Mac
users start automatically).
Windows 7 or 8 users enter the command below (substituting your login ID) o get to your home directory:
cd C:\Users\yourLoginId

The cd is short for “Change Directory”, changing the current directory.

Warning: Windows only: The cd command does not work the way you are likely to think about it on a Windows
system with more than one drive (like C: and flash drive E: that you have plugged in). Windows remembers a
separate current directory for each separate drive. It also separately remembers a current drive. You do not change
the current drive with the cd command. The command to change the current drive is just the name of the drive with
a colon after it. For example the command

E:

sets the current drive to E:, and the active directory is the current directory on E:.
However, if the current drive is C:, and you enter the command

cd E:\comp170

then you change the current directory on E:, but the current drive remains C:.

Suppose you don’t know the path to your hello directory on Windows, but can you can find it in an Windows Explorer
window (right clicking on Start); here’s how to provide that path to the cd command:

18.3. Command Line Introduction 203

Introduction to Computer Science in C#, Release 2.0

• Depending on the setup of your options, in the address bar you may not see a clear path with a drive and
backslashes. In that case generally clicking to the right of any directory in the path converts the view to the
version we use on the command line.
• When you see a full absolute path, you can just note it and manually copy it, or select it all and copy it, and
follow the instructions in Copy and Paste to later paste in the command window.
• In any case click in the Mono Command Prompt window, type cd and a space, then type or paste the path.
• Of course, you can also go the other way – if you see the current directory name in the Windows prompt, type
that into an Explorer address bar to see its contents in a GUI window
On a Mac there is an easier shortcut:
• Type cd and a space to start the command in the terminal
• Locate the directory you want in the Finder (not opening the directory).
• Drag the directory icon to the terminal. The path gets pasted! Press return.

18.3.2 Common Commands

The command shell is now waiting for you to type in a command (a short name that the shell recognizes) followed by
0 or more parameters separated by spaces (and Enter). Note that if a parameter contains spaces you must surround the
parameter value with matching single or double quotes – you’ll see an example later.
We are going to mention some of the simplest uses of basic commands. More advanced documentation would include
more options.
Some commands are common between the Windows and Mac shells:
dir or ls to list all the files a in the current directory or a named directory.
cd stands for Change Directory – you can use this command to change the current working directory to a different
one.
You can use this command to change to directories where your C# program source files are located, if different
from the initial directory.
On Windows, suppose you created a directory C:\COMP170\hello; to change to that, type cd C:\COMP170\hello
and press Enter – the shell prompt will change to show this new directory location and programs like mcs and
mono will be able to “see” (access) files there, directly by name. If the Comp170 directory was you current
directory, it would be shorter to use relative paths and just cd hello.
On a Mac you can also use either an absolute or a relative path with cd.
If you included a space in one or more of the directory names, for example C:\COMP 170\hello (a space between
COMP and 170) you should enclose that part(s) in quotes like so: cd C:\”COMP 170”\hello
Mac Note: if you type just cd and press Enter you will change back to your home directory. There is also a
shorthand name for your home directory in command paths: tilde (~), often shifted backquote. Sorry, no such
thing with Windows.
mkdir stands for make directory – you can use mkdir to create a new empty directory in the current directory.
For example, on a Mac with current directory /Users/YourLogin, type mkdir hello and press Enter – this will
create a new directory /Users/yourLogin/hello if it did not exist before; you can now create a C# source file in
that directory and enter cd hello in the command shell.
An optional Windows abbreviation is md.
rmdir removes an empty directory that you give as parameter, e.g.,
rmdir hello

204 Chapter 18. Appendix

 Introduction to Computer Science in C#, Release 2.0

Then, with Mono installed (and for Windows, with a Mono command window), the programs associated with Mono
can be used:
mcs compiles one or more listed C# source files without using Xamarin Studio.
csharp is the interactive C# statement testing program.
Other useful commands window commands with different names for Windows and Mac, listed by generic function,
with general Windows syntax first and Mac second, and then often examples in the same order:
Display the contents of a text file in the command window. Name origin: a more complicated use of cat is to
concatenate files.
type textFileName
cat textFileName

type my_program.cs
cat my_program.cs

Make a copy of a file. Caution: If the second file already exists, you wipe out the original contents!
copy originalFile copyName
cp originalFile copyName

copy prog1.cs prog2.cs

cp prog1.cs prog2.cs

Erase or remove a file:

erase fileToKill
rm fileToKill

erase poorAttempt.cs
rm poorAttempt.cs

Another Windows equivalent is del (short for delete).

Help on a command:
help commandName;
commandName –help
Note the double dash above: This sometimes works for concise help on a Mac while you can generally get immensely
detailed help overload on a Mac from
man commandName

18.3.3 Scripts

This is not a subject of this course, but commands can be combined into script files.

18.3. Command Line Introduction 205

Introduction to Computer Science in C#, Release 2.0

Scripting languages are in fact whole new specialized programming languages, that include many of the types of
programming statements found in C#.

18.3.4 Copy and Paste

Copying or pasting with a Mac is is the same with a terminal as in other editing: Use the same Apple Command key
with C or P, and you can select with the mouse.
In Windows it is more complicated to use a command window: You can paste into the current command line by right
clicking on the Command Window Title bar, and select edit and then paste.
By default a Windows command window is not sensitive to the mouse. You can change so that it is sensitive for select
and copy: Right click in the title bar, select defaults, and make sure the check boxes under edit options are all checked.
(The last two are explained in the next section.) Click OK. Then you can select with mouse and press Enter for the
selection to be remembered in the copy buffer.

18.3.5 Command Line Shortcuts

Both Windows and Mac (with the right options selected, like the Windows check boxes in the last section), allow you
to reduce typing:
You can bring back a previous command for the history of commands that are automatically remembered: Use the up
and down arrows. This makes it very easy to run the same command again, or to make slight edits.
Both Windows and OS-X can see what files are in any directory being referred to. If you just start to type a file or
folder name and then press the Tab key, both Windows and OS-X will do file completion to complete the name if there
is no ambiguity. If there is ambiguity, they work differently:
• Windows will cycle through all the options as you keep pressing Tab.
• On the first tab OS-X will do nothing but give a sound if there is ambiguity, but the second tab will list all the
options. Then you need to type enough more to disambiguate the meaning.

18.4 Precedence of Operators

Earlier lines have higher precedence. Only operators used in this book are included:
obj.field f(x) a[i] n++ n-- new
+ - ! (Type)x (Unary operators)
* / %
+ - (binary)
< > <= >=
== !=
&&
||
= *= /= %= += -=

Grouping parentheses are encouraged with less common combinations, even if not strictly necessary.

18.5 Homework: Grade Calculation

You are going to be putting together your first programming assignment where you will be taking the various concepts
we have learned thus far in lecture and lab and to put together your first meaningful program on your own.

206 Chapter 18. Appendix

 Introduction to Computer Science in C#, Release 2.0

This program will incorporate the following elements:

• Prompt a user for input.
• Perform some rudimentary calculations.
• Make some decisions.
• Produce output.
As we’ve mentioned in the early lectures, our focus is going to be on learning how to write computer programs that
start with a Main() function and perhaps use other functions as needed to get a particular job done. Eventually, we
will be incorporating more and more advanced elements, such as classes and objects. For now, we would like you to
organize your program according to the guidelines set forth here.

18.5.1 Program Summary

Our first program is based on a common task that every course professor/instructor needs to do: make grades. In any
given course, there is a grading scale and a set of categories.
Here is sample output from two runs of the program. The only data entered by the user are show in boldface for
illustration here.
One successful run with the data used above:
Enter weights for each part as an integer
percentage of the final grade:
Exams: 40
Labs: 15
Homework: 15
Project: 20
Participation: 10

Enter decimal numbers for the averages in each part:

Exams: 50
Labs: 100
Homework: 100
Project: 100
Participation: 5

Your grade is 70.5%

Your letter grade is C-.
A run with bad weights:
Enter weights for each part as an integer
percentage of the final grade:
Exams: 30
Labs: 10
Homework: 10
Project: 10
Participation: 10

Your weights add to 70, not 100.

18.5. Homework: Grade Calculation 207

Introduction to Computer Science in C#, Release 2.0

This grading program is ending.

18.5.2 Details

This is based on the idea of Dr. Thiruvathukal’s own legendary course syllabus. We’re going to start by assuming that
there is a fixed set of categories. As an example we assume Dr. Thiruvathukal’s categories.
In the example below we work out for Dr. Thiruvathukal’s weights in each category, though your program should
prompt the user for these integer percentages:
• exams - 40% (integer weight is 40)
• labs - 15% (weight 15)
• homework - 15% (weight 15)
• project - 20% (weight 20)
• participation - 10% (weight 10)
Your program will prompt the user for each the weights for each of the categories. These weights will be entered as
integers, which must add up to 100.
If the weights do not add up to 100, print a message and end the program. You can use an if-else construction
here. An alternative is an if statement to test for a bad sum. In the block of statements that go with the if statement,
you can put not only the message to the user, but also a statement:
return;

Recall that a function ends when a return statement is reached. You may not have heard that this can also be used with
a void function. In a void function there is no return value in the return statement.
Assuming the weights add to 100, then we will use these weights as floating point numbers to compute your grade.
We will be using double, which gives you the best precision when it comes to floating-point arithmetic.
We’ll talk in class about why we want the weights to be integers. Because floating-point mathematics is not 100%
precise, it is important that we have an accurate way to know that the weights really add up to 100. The only way to
be assured of this is to use integers. We will actually use floating-point calculations to compute the grade, because we
have a certain tolerance for errors at this stage. (This is a fairly advanced topic that is covered extensively in courses
like COMP 264/Systems Programming and even more advanced courses like Numerical Analysis, Comp 308.)
We are going to pretend that we already know our score (as a percentage) for each one of these categories, so it will
be fairly simple to compute the grade.
For each category, you will define a weight (int) and a score (double). Then you will sum up the weight * score and
divide by 100.0 (to get a double-precision floating-point result).
This is best illustrated by example.
George is a student in COMP 170. He has the following averages for each category to date:
• exams: 50%
• labs: 100%
• homework: 100%
• project: 100%
• participation: 5%
The following session with the csharp interpreter shows the how you would declare all of the needed variables and
the calculation to be performed:

208 Chapter 18. Appendix

 Introduction to Computer Science in C#, Release 2.0

csharp> int exam_weight = 40;

csharp> int lab_weight = 15;
csharp> int hw_weight = 15;
csharp> int project_weight = 20;
csharp> int participation_weight = 10;

csharp> double exam_grade = 50.0;

csharp> double lab_grade = 100;
csharp> double homework_grade = 100;
csharp> double project_grade = 100;
csharp> double participation_grade = 5;

csharp> ShowVars();
int exam_weight = 40
double lab_weight = 15
int hw_weight = 15
int project_weight = 20
int participation_weight = 10
double exam_grade = 50
double homework_grade = 100
double lab_grade = 100
double project_grade = 100
double participation_grade = 5

This is intended only to be as an example though. Your program must ask the user to enter each of these variables.
Once we have all of the weights and scores entered, we can calculate the grade as follows. This is a long expression:
It is continued on multiple lines. Recall all the > symbols are csharp prompts are not part of the expression:
csharp> double grade = (exam_weight * exam_grade +
> homework_weight* homework_grade +
> lab_weight * lab_grade + project_weight * project_grade +
> participation_weight * participation_grade) / 100.0;

Then you can display the grade as a percentage:

csharp> Console.WriteLine("Your grade is {0}%", grade);
Your grade is 70.5%

Now for the fun part. We will use if statements to print the letter grade. You will actually need to use multiple if
statements to test the conditions. A way of thinking of how you would write the logic for determining your grade is
similar to how you tend to think of the best grade you can hope for in any given class. (We know that we used to do
this as students.)
Here is the thought process:
• If my grade is 93 (93.0) or higher, I’m getting an A.
• If my grade is 90 or higher (but less than 93), I am getting an A-.
• If my grade is 87 or higher (but less than 90), I am getting a B+.
• And so on...
• Finally, if I am less than 60, I am unlikely to pass.
We’ll come to see how logic plays a major role in computer science–sometimes even more of a role than other math-
ematical aspects. In this particular program, however, we see a bit of the best of both worlds. We’re doing arithmetic
calculations to compute the grade. But we are using logic to determine the grade in the cold reality that we all know
and love: the bottom-line grade.

18.5. Homework: Grade Calculation 209

Introduction to Computer Science in C#, Release 2.0

This assignment can be started after the data chapter, because you can do most all of it with tools learned so far. Add
the parts with if statements when you have been introduced to if statements. (Initially be sure to use data that makes
the weights actually add up to 100.)
You should be able to write the program more concisely and readably if you use functions developed in class for the
prompting user input.

18.5.3 Grading Rubric

Warning: As a general rule, we expect programs to be complete, compile correctly, run, and be thoroughly tested.
We are able to grade an incomplete program but will only give at most 10/25 for effort. Instead of submitting
something incomplete, you are encouraged to complete your program and submit it per the late policy. Start early
and get help!

25 point assignment broken down as follows:

• Enter weights, with prompts [3]
• End if the weights do not add to 100: [5]
• Enter grades, with prompts: [3]
• Calculate the numerical average and display with a label: [5]
• Calculate the letter grade and display witha label: [5]
• Use formatting standards for indentation: [4]
– Sequential statements at the same level of indentation
– Blocks of statements inside of braces indented
– Closing brace for a statement block always lining up with the heading before the start of the block.

18.5.4 Logs and Partners

You may work with a partner, following good pair-programming practice, sharing responsibility for all parts.
Only one of a pair needs to submit the actual programming assignment. However both students, independently, should
write and include a log in their Homework submission. Students working alone should also submit a log, with fewer
parts.
Each individual’s log should indicate each of the following clearly:
• Your name and who your partner is (if you have one)
• Your approximate total number of hours working on the homework
• Some comment about how it went - what was hard ...
• An assessment of your contribution (if you have a partner)
• An assessment of your partner’s contribution (if you have a partner).
Just omit the parts about a partner if you do not have one.

Note: Name the log file with the exact file name: “log.txt” and make it a plain text file. You can create it in a program
editor or in a fancy document editor. If you use a fancy document editor, be sure to a “Save As...” dialog, and select
the file format “plain text”, usually indicated by the ”.txt” suffix. It does not work to save a file in the default word
processor format, and then later just change its name (but not its format) in the file system.

210 Chapter 18. Appendix

 Introduction to Computer Science in C#, Release 2.0

18.6 Homework: Grade Calculation from Individual Scores

In the previous assignment, we calculated grades based on a memorized overall grade within each of the categories
below, as in this example:
• exams - 40% (integer weight is 40)
• labs - 15% (weight 15)
• homework - 15% (weight 15)
• project - 20% (weight 20)
• participation - 10% (weight 10)
In this assignment, we are going to change the specification slightly to make the program a bit smarter. Instead of
someone having to remember what their average grade was for each category, we will prompt the user for the number
of items within each category (e.g. number of exams, number of labs, etc.), have the user enter individual grades, and
have the program calculate the average for the category.
As usual, we will begin by specifying requirements. User responses are shown bold faced.

18.6.1 Functional Requirements

1. Instead of bombing out if the weights don’t add up to 100, use Do-While Loops to prompt the user again for all
of the weights until they do add up to 100. A do { ... } while loop is the right choice here, because
you can test all of the weights at the end of the loop, after each time they have been entered in the loop.
2. Write a function, FindAverage, to do the following. The example refers to the category exam, but you
will want your code to work for each category, and hence the category name will need to be a parameter to
FindAverage.)
Prompt the user for the number of items in the category:
Please enter the number grades in category exam: 4
Instead of prompting the user for an overall average exam grade, use a loop to read one grade at a time. The
grades will be added together (on the fly) to give the grade for that category. For example, after you have asked
for the number of exams, you’d prompt the user to enter each exam grade and have the program compute the
sum. To make sure everyone understands what should be happening, you should also print a running total of
the grade category you’re calculating:
Please enter the grade for exam 1: 100

Please enter the grade for exam 2: 90

Please enter the grade for exam 3: 80

Please enter the grade for exam 4: 92

Calculated average exam grade = 90.5

Of course you must return the grade to the caller for use in the overall weighted average grade.
A category may have only a single grade, in which case the user will just enter the number of grades as 1.
3. Once you have read in the data for each of the items within a category, you’ll basically be able to reuse the code
that you developed in the previous assignment to compute the weighted average and print the final letter grade.

18.6. Homework: Grade Calculation from Individual Scores 211

Introduction to Computer Science in C#, Release 2.0

4. Print the final numerical average rounded to one decimal place. If the final average was actually 93.125, you
would print 93.1. If the final average was actually 93, you would print 93.0. If the final average was actually
93.175, you would print 93.2.

18.6.2 Style Requirements

1. For this assignment, you are expected to start using functions for all aspects of the assignment. For example, it
can become tedious in a hurry to write code to prompt for each of exams, labs, homework, etc. when a single
function (with parameter named category) could be used to avoid repeating yourself. In particular you should
write your function to take advantage of our UI class, from User Input: UI.
2. Also beginning with this assignment, it is expected that your work will be presented neatly. That is, we expect
the following:
• proper indentation that makes your program more readable by other humans. Use all spaces, not tabs to
indent. You never know what default tabs your grader will have set up.
• proper naming of classes and functions. In C#, the convention is to begin a name with a capital letter. You
can have multiple words in a name, but these should be capitalized using a method known as CamelCase
[CamelCase]. We also recommend this same naming convention for variables but with a lowercase first
letter. For variables, we are also ok with the use of underscores. For example, in homework 1 we used
names like exam_grade. If you use CamelCase, you can name this variable examGrade.
• If you have any questions about the neatness or appearance of your code, please talk to the instructor or
teaching assistant.
• This guide from CIS 193 at [UPennCSharp] provides a nice set of conventions to follow. We include this
here so you know that other faculty at other universities also consider neatness/appearance to be important.

18.6.3 Grading Rubric

Warning: As a general rule, we expect programs to be complete, compile correctly, run, and be thoroughly tested.
We are able to grade an incomplete program but will only give at most 10/25 for effort. Instead of submitting
something incomplete, you are encouraged to complete your program and submit it per the late policy. Start early
and get help!

25 point assignment broken down as follows:

• Loop until weights add to 100: 5
• Average any number of grades in a category: 5
• One function that is reused and works for the average in each category: 5
• Print final numerical grade rounded to one decimal place: 2
• Previous program features still work: 3
• Style: 5

18.6.4 Logs and Partners

You may work with a partner, following good pair-programming practice, sharing responsibility for all parts.

212 Chapter 18. Appendix

 Introduction to Computer Science in C#, Release 2.0

Only one of a pair needs to submit the actual programming assignment. However both students, independently, should
write and include a log in their Homework submission. Students working alone should also submit a log, with fewer
parts.
Each individual’s log should indicate each of the following clearly:
• Your name and who your partner is (if you have one)
• Your approximate total number of hours working on the homework
• Some comment about how it went - what was hard ...
• An assessment of your contribution (if you have a partner)
• An assessment of your partner’s contribution (if you have a partner).
Just omit the parts about a partner if you do not have one.

Note: Name the log file with the exact file name: “log.txt” and make it a plain text file. You can create it in a program
editor or in a fancy document editor. If you use a fancy document editor, be sure to a “Save As...” dialog, and select
the file format “plain text”, usually indicated by the ”.txt” suffix. It does not work to save a file in the default word
processor format, and then just change its name (but not its format) in the file system.

18.7 Homework: Grade File

Copy project files in grade_file_homework_stub to your own project. Then you should have copies of the source
file grade_files.cs for you to complete for this homework. The folder also contains sample data files including the
examples discussed below.
In this assignment, we’re going to begin taking steps to help you achieve greater independence when it comes to
programming. This means (among other things) that you will be given what is commonly known as a specification. In
software development–and in the business world in general, it is customary to capture a set of business requirements
in what is commonly known as a requirements specification document. While what you read here will be much more
concise, we want you to become familiar with requirements-driven thinking, without which many real-world software
projects fail.
After presenting the set of requirements, we will give you some hints for how to implement the requirements. These
hints may or may not prove completely helpful to you, and you are also invited to come up with your own solutions.
As we inch closer to the semester project, you’re going to want to use your imagination to create a good solution to a
problem.

18.7.1 Brief Problem Statement

The previous two homework assignments represent a great simplification of the real-world process of grading. The
notion that grade information must be entered manually is rather tedious, not to mention error prone. In the real world,
grade information would be kept in a file (a spreadsheet is common), from which various calculations and summary
reports could be generated.
In this assignment, the problem we are trying to solve is to take all of the raw grade data from one or more student
files and prepare a summary report file with a line for each student.
Although we could do all of what we’re describing here with a spreadsheet, the point is to show how we can use C# to
read in a simplified form of comma-separated data, process it, and do some general-purpose calculations on the data.

18.7. Homework: Grade File 213

Introduction to Computer Science in C#, Release 2.0

18.7.2 Using C#

We’ll be making use of a number of C# features (some old, some new) in this homework:
• decisions, loops, strings, and functions
• files
• arrays

18.7.3 Requirements

1. Unlike in previous assignments, this program must accept data from a collection of input files (that is, it will not
be reading most of the data from the class Console).
2. The program needs a course abbreviation from the user. If there is a command line argument, use it as the
course abbreviation. Make sure your code can read a command-line argument using the special form of
Main(string[] args) already in the stub grade_files.cs. If the user does not provide at command line
argument, prompt the user for it once the program starts. The abbreviation should not include spaces. An ex-
ample would be comp170. All data files will include the course abbreviation as part of their name. We will use
comp170 in the examples below, but it could be something else. The folder also contains sample data files for a
course abbreviation comp150.
Note that these data files are not in the Xamarin Studio execution directory, but in the project directory, so the
FIO Helper Class is useful to provide flexibility in reading the data files.
3. There are two master files for any course. One is “categories_” + the course abbreviation + ”.txt”. For example,
categories_comp170.txt is a sample data file provided and used below.
It will contain three lines. The first line is a comma separated list of category names like
There may be extra spaces after the commas. Categories will be chosen so that each one starts with a different
letter.
The second line contains the integer weights for each category, like
They do not need to add to 100. If the sum is called totWeights, get the final grade by summing for each
category:
(category weight)(category grade)/totWeights

The third line will contain the number of grades in each category, like
The second master file will be “students_” + the course abbreviation + ”.txt”. For example
students_comp170.txt. It will contain a list of student information records. Each record (one per in-
put line) will have the following structure:
Student ID, Last Name, First Name
For example, the sample data file students_comp170.txt starts:
4. There will be a secondary file for each student, named after the student id and the course abbreviation and
”.data”. For example, John’s scores would be kept in a file named P12345678comp170.data. Maria’s
scores would be in P00000001comp170.data. Each record (one per file line will have the following struc-
ture:
Category letter, Item, Points Earned
where:
• category letter is the first letter of the category. With the categories given in the example above, they would
be E, L, H, P, and C.

214 Chapter 18. Appendix

 Introduction to Computer Science in C#, Release 2.0

• item is a number within that category (1, 2, 3, ...) - only used in part of the extra credit.
• points earned is a real number
• the lines are in no special order.
Sample data file P12345678comp170.data starts:
5. The program will process the data from each student file and calculate the average within each category, and
then the weighted overall average. Also display the letter grade for each student, using code derived from the
previous assignment.
6. Your program writes the final report file. It is named with the course abbreviation + “_summary.txt”. Example:
“comp170_summary.txt”. This file must have a line for each student showing the student’s last name, first name,
weighted average rounded to one decimal place, and letter grade. File comp170_summary.txt would start
with lines:
Doe, John 78.9 C+
Hernandez, Maria 88.2 B+

Write this file to the same directory where you found the input data. Again the FIO Helper Class is useful.
7. The rest of the test data for course abbreviations comp170 and all the data for comp150 in the homework
directory. There are also sample solution files for the summaries (including some extra credit additions at the
ends of lines). Their names end in _solution.txt to distinguish them from the summary files you should
generate in tests.
While your program should certainly work for course abbreviations comp170 and comp150, it should also work
in general for any data files your refer to in the defined formats and place in the same folder.

18.7.4 Hints

1. Read Files, Paths, and Directories. You’re still going to need ReadLine() and WriteLine() in this assignment,
the only difference is that we’ll get the input from a file instead of the Console. The parameter syntax will be
the same.
2. For each file line you’ll want to use the String Method Split, and then the Trim method from More String
Methods on each part to remove surrounding spaces. Then use indexing to get the field of interest. (More
below.)
3. You’ll need an outer loop to read the records from the master name file. You’ll need an inner loop (or a loop
inside of a function) to read the records for each student.
4. When processing the records from a student file, you should process each one separately and not assumed they
are grouped in any particular order.
This means, specifically, that your program simply reads a record, decides what category it is in, and updates
the running total for that category. Once the entire file has been read, you can compute the average for each
category based on the number of items that should be in that category, which may be more than the number of
records in the file for items turned in.
5. There is no need to keep a score after you’ve read it and immediately used it. Do use an array, however, for the
running total for each category.
6. In order to deal with a varying number of categories and different possible first letter codes, you will need to
split the category name line into an array, say
string[] categories;

To know where to update data for each category, you can use this function after you read in a code, to determine
the proper index. It is already in the stub of the solution file grade_files.cs:

18.7. Homework: Grade File 215

Introduction to Computer Science in C#, Release 2.0

You may assume the data is good and the -1 is never returned, but the compiler requires this last line.
7. You cannot have one fixed formula to calculate the final weighted grade, because you do not know the number
of categories when writing the code. You will have to accumulate parts in a loop.
8. Test thoroughly! Be sure to test with and without command line parameter and with multiple data sets.

18.7.5 Grading Rubric (25 points)

1. Get the abbreviation from the command line if it is there. [2]

2. Otherwise get the abbreviation from prompting the user. [1]
3. Read the categories file and parse lines. [2]
4. Deal with each student. [3]
5. Calculate the cumulative grades in each category, reading a student’s file once, using arrays. [5]
6. Calculate the overall grade and letter grade. [3]
7. Generate summary entries. [3]
8. Use functions where there would otherwise be two several-line blocks of code differing only in the name of the
data evaluated and the name of the result generated. [2]
9. Use good style: formatting, naming conventions, meaningful names other than for simple array indices, lack of
redundant code. [4]
Optional Extra Credit Opportunities! You may choose to do any combination that does not include both of the last
two options about missing work.
1. Format the summary file in nice columns. Include the grades for each category, rounded to one decimal place.
Include a heading line. For example the summary for the repository example Comp150 could start:
Name: Last, First Avg Gr E H P
Hopper, Grace 100.0 A 100.0 100.0 100.0

You may assume the last-first name field fits in 25 columns. Copy the first three column headings from above.
The column headings for the categories can just be their one letter code. Names and letter grades should be
left-justified (padded on the right, by using a negative field width). See Left Justification. [2]
2. Change the scheme for calculating letter grades to use a function that calculates the proper grade, where the only
if statement is one simple one inside a loop. The if statement will have a return statement in its body, and
no else. The loop will need to use corresponding arrays of data for grade cutoffs and grade names. [3]
3. For any student who has missed passing in all the required items, generate extra data on missing work in the
summary, at the right end of the line for the student. Add this to whichever version of the earlier parts you use.
Include an addendum starting with “Missing: ” only if there are not enough grades in one or more categories.
For each category where one or more grades is missing, including a count of the number of grades missing
followed by the category letter. An example using the example categories is:
Doe, John 68.5 D+ Missing: 2 L 1 H
Smith, Chris 83.2 B Missing: 1 L
Star, Anna 91.2 A-

meaning Doe has 2 labs missing and 1 homework missing. Smith is missing one lab. Star has done all assigned
work, since nothing is added. The solution files display this extra credit addition on the ends of lines. [3]
4. This is a much harder alternate version for handling missing work: Unlike the previous format, do not count
and print the number of missing entries in each category in a form like “2 L ”. Replace such an entry with a list
of each item missing, in order, as in “L:1, 4 ”, meaning labs 1 and 4 were missing. Assume that the expected

216 Chapter 18. Appendix

 Introduction to Computer Science in C#, Release 2.0

item numbers for a category run from 1 through the number of grades in the category. You may assume no item
number for the same category appears twice. For example, with the sample data files given in the repository for
comp170, the summary line for John Doe would be:
Doe, John 78.9 C+ Missing: L: 1, 4 H: 3

The most straightforward way to do this requires something like an array of arrays, array of lists or array of sets.
You may need to read ahead if you want to use one of these approaches. [5]

18.8 Homework: Book List

Objectives:
• Complete a simple data storing class (Book), with fields for title, author, and year of publication.
• Complete a class with a Collection (BookList) that uses the public methods of another class you wrote (Book),
and select various data from the list.
• Complete a testing program (TestBookList), that creates a BookList, adds Books to the BookList, and tests
BookList methods clearly and completely for a user looking at the output of the program and not the source
code..
Copy stub files from the project books_homework_stub to your own project. Stubs for the assignment files are book.cs,
book_list.cs, and test_book_list.cs,
Some of the method stubs included are only to be fleshed out if you are doing the corresponding extra credit option.
They include a comment, just inside the method, // code for extra credit. There are also extra files used
by the extra credit portion. They are discussed in the Extra Credit section at the end of the assignment.
Complete the first line in each file to show your names. At the top of the Book class include any comments about help
in all of the classes.
Create methods one at a time, and test them. Complete book.cs first, preferably testing along the way. (You can write
an initial version of the testing program, so it does not depend on BookList.) Then add methods to book_list.cs,
and concurrently add and run tests in test_book_list.cs. Testing the Book class first means that when you get
to BookList you can have more confidence that any problems you have are from the latest part you wrote, not parts
written earlier in the class Book.
Remember to have each individual submit a log, log.txt, in the same format as the last assignment.

18.8.1 Book class

See the stub file provided. It should have instance fields for the author, title, and year (published).
Complete the constructor:
public Book(string title, string author, int year)

that initializes the fields. (Be careful, as we have discussed in class, when using the same names for these parameters
as the instance variables!)
It should have three standard (one line) getter methods:
public string GetTitle()

public string GetAuthor()

public int GetYear()

18.8. Homework: Book List 217

Introduction to Computer Science in C#, Release 2.0

and
public override string ToString()

ToString should return a single string spread across three lines, with no newline at the end. For example if the
Book fields were “C# Yellow Book”, “Rob Miles”, and 2011, the string should appear, when printed, as
Title: C# Yellow Book
Author: Rob Miles
Year: 2011

The override in the heading is important so the compiler knows that this is the official method for the system to
used implicitly to convert the object to a string.
Remember the use of @ with multi-line string literals.

18.8.2 BookList class

It has just one instance variable, already declared:

private List<Book> list;

It has a constructor (already written - creating an empty List):

public BookList()

It should have public methods:

The regular version should just leave the final return true; The extra credit version is more elaborate.
Further methods:
For instance if the list included books published in 1807, 1983, 2004, 1948, 1990, and 2001, the statement
PrintBooksInYears(1940, 1990);

would list the books from 1983, 1948, and 1990.

18.8.3 TestBookList class

It should have a Main program that creates a BookList, adds some books to it (more than in the skeleton!), and
convincingly displays tests of each of BookList’s methods that exercise all paths through your code. Check for one-off
errors in PrintBookYears. With all the methods that print something, the results are easy to see. Do print a label, as in
the skeleton, before printing output from each method test, so that the user of the program can see the correctness of
the test without any knowledge of the source code!

18.8.4 Grading Rubric

Book class. Requires the constructor. Then

• [1 point] public Book(string title, string author, int year)
• [1] public string GetTitle()
• [1] public string GetAuthor()
• [1] public int GetYear()
• [2] public override string ToString()

218 Chapter 18. Appendix

 Introduction to Computer Science in C#, Release 2.0

BookList class
• [2] public bool AddBook(Book book)
• [2] public void PrintList()
• [2] public void PrintTitlesByAuthor(string author)
• [2] public void PrintBooksInYears(int firstYear, int lastYear)
TestBookList
• [2] Supply data to screen indicating what test is being done with what data and what results, so it is clear that
each test works without looking at the source code.
• [5] Convincingly display tests of each of BookList’s methods that exercise all paths through your code.
Overall:
• [4] Make your code easy to read - follow indenting standards, use reasonable identifier names.... Do not duplicate
code when you could call a method already written.

18.8.5 Extra Credit

You may do any of the numbered options, except that the last one requires you to do the previous one first.
To get full credit for any particular option, tests for it must be fully integrated into TestBookList!
1. [2 points] Complete
Also change the PrintList method body to the one line:
Console.Write(this);

(The Write and WriteLine methods print objects by using their ToString methods.)
Be sure to make this addition to TestBookList: Test the ToString method by converting the resulting BookList
description string to upper case before printing it (which should produce a different result than the regular mixed
case of the PrintList method test).
2. [4 points]
In the Book class, a new constructor:
In class BookList, a new constructor:
For testing we included special files in the right format: books_homework_stub/books.txt and
books_homework_stub/morebooks.txt.
You will also want to include a reference to fio/fio.cs, so the text files are easy to find.
3. [4 points]
In class Book:
It is essential to have the IsEqual method working in Book before any of the new code in BookList, which all
depends on the definition of IsEqual for a Book.
NOTE: We chose the name IsEqual to distinguish it from the more general Equals override that you could
write. The Equals override allows for a parameter of any object type. With skills from Comp 271 you you be
able to write the Equals override.
In class BookList:

18.8. Homework: Book List 219

Introduction to Computer Science in C#, Release 2.0

Caution: Do NOT try to use the List method Contains: Because we did not override the Equals method
to specialize it for Books, the List method Contains will fail. You need to do a litle bit more work and write
your own version with a loop.
Change the AddBook method from the regular assignment, so it satisfies this documentation:
In TestBookList you need to react to the return value, too.
4. [2 points] This one requires the previous elaboration of AddBook. In BookList:
You might want to code it first without worrying about the correct return value; then do the complete version.
There are multiple approach to determining the return value, some much easier than others!
To fully test in TestBookList, you need to react to the return value, too.

18.9 Lab: Version Control

Modern software development requires an early introduction source code management, also known as version control.
While source code management is frequently touted as being beneficial to a project team, it is also of great value for
individuals and provides a clear mechanism for tracking, preserving, and sharing your work.
In addition, it simplifies the process of demonstrating and submitting your work to your instructor (and graduate
assistants). Long gone are the days of carrying stuff around on USB drives and sending e-mail attachments.
There are numerous options for version control. In the free/open source community, a number of solutions have
emerged, including some old (CVS, Subversion) and some new (Mercurial, Git, and Bazaar). We’re particularly fond
of the Mercurial system. A key reason for our choice is that there is an excellent cloud-based solution to host your
projects known as Bitbucket (http://bitbucket.org).
Mercurial is set up for our labs. You can install it for your personal machine from
http://mercurial.selenic.com/downloads/ .
There are other similar solutions to Bitbucket but none at present provides a completely free solution for hosting
private repositories, which allow you to keep your work secret from others.
The basic idea is to keep a main current copy of a project at a place like bitbucket. Anywhere that you work, you can
download a copy of the central version. You can add and change files. There are several layers insulating changes to
local files from changes to the central repository:
• You must explicitly add any new file names you want the repository to track.
• Even on a tracked file, you must commit changes to the local repository.
• For the committed changes to get to the central repository, you must push them.
• You have control over what files get ignored.
Later, when you want the latest changes to the central repository to get to your site, you need to pull data from
the central repository, and then explicitly update your local repository, to incorporate the new data from the central
repository.

18.9.1 Goals

In this lab, we’re going to learn:

1. How to create a source code repository.
2. How to add files and folders to a repository and track them.
3. How to make sure certain files and folders are not kept in the repository.

220 Chapter 18. Appendix

 Introduction to Computer Science in C#, Release 2.0

4. How to push your changes to a remote repository (at Bitbucket.org).

5. How to do your work at home and in the lab.
6. How to get our book, examples, and projects (now that that you know how to use Mercurial).

Before We Proceed

You should know that this lab is designed to be repeated as many times as needed. You can create as many repositories
as you wish, subject to the limitations at Bitbucket.org, and may want to create different repositories for different needs
(one for yourself, one for you and your partner in pair programming, etc.) This lab assumes you are starting with a
repository for your own work. We’ll include a step for how to add a collaborator to the project.
Future labs will talk about additional things you need to know when it comes to collaboration, so please view this lab
as a beginning aimed at helping you to start using version control right away for your own projects.

18.9.2 Steps

Create Bitbucket.org account

We’re going to begin by creating a remote repository for our work. The advantage of doing so is that we get a hosted
repository that we can use to push/pull our work. (Unlike a dangerous stunt, you want to be able to try this at home,
too!)
Signing up for a repository at http://bitbucket.org is easy. From the landing page, just click on the option for the Free
Plan. This allows you to create any number of public/private repositories with support for up to 5 users. This is all
you’ll need for your work in this course.
Once you’ve created your account (and confirmed it, if required) you are good to go for the rest of this lab!

Create repository at Bitbucket

Now we’ll create a first repository at Bitbucket.org.

Go to Repositories -> Create Repository (the option is at the bottom of the list of menu options). You’ll
see this screen:

18.9. Lab: Version Control 221

Introduction to Computer Science in C#, Release 2.0

You’ll need to fill in or select the following options:

• Name: A short name for your project. You are encouraged to keep this simple. If you are using this for all of
your work in COMP 170 (which is fine) you might name the repository after your initials. So if your name is
Linus Torvalds, you could give a short name like LinusTorvaldsCOMP170 or LTCOMP170.
• Repository Type: Select Mercurial. Yes, we realize that Xamarin Studio supports Git natively, but for the reasons
mentioned earlier, we have chosen Mercurial. We will allow you to use Git on your own if you can figure it out
and use it properly. But this lab assumes Mercurial.
• Language: You can select anything you like here. We do C# for the most part in this class, so we recommend
that you select it.
• Description: You can give any description you like. If you are working with a partner please list both you and
your partner’s name in the description.
• Web Site: Optional
• Private checkbox should be checked.
So just go ahead and create your first repository. You can always create more of them later.
Here is an example of a filled out form:

222 Chapter 18. Appendix

 Introduction to Computer Science in C#, Release 2.0

Set up your Mercurial commit username

If you in a place where you have a permanent home directory, like on your machine or in the Linux Lab, create a
file named .hgrc in your home directory. Your home directory is where you are dumped when you open a DOS or
Linux/OS X terminal. This is not inside your repository. This file must contain the following lines, with the part after
the equal sign personalized for you:
[ui]
username = John Doe <[email protected]>

It is a convention to give a name and email address, though it does not need to match the email address you gave when
signing up for bitbucket.
Creating this file saves you the trouble of having to pass the -u username option to hg each time you do a commit
operation.
You can put this file in your home directory in Windows labs, but it disappears. You might want to keep an extra copy
in your repository, and copy it to the Windows home folder when in the lab.
As a gentle reminder, your home directory on Windows can be a bit difficult to find. The easiest way is to use your
editor to locate your home folder. When in the DOS prompt, you will also see the path to your directory as part of the
prompt. For example, on Windows 7, you will see C:\Users\johndoe.

Warning: To ensure that you did this step correctly, please open a new terminal or DOS window at this time and
use the ls or dir command to verify that the .hgrc file is indeed present in your home directory. If it is in any
other folder, Mercurial (the hg command) will not be able to find it–and you will receive an error.

18.9. Lab: Version Control 223

Introduction to Computer Science in C#, Release 2.0

Clone a repository from Bitbucket

Open a terminal or DOS command shell.

On Windows, the Mono shell is not appropriate. You can get a regular DOS command shell by clicking the start menu
and typing cmd and into the text box at the bottom of the start menu, and pressing return.
In the terminal/DOS-shell navigate with to the the directory where you want to place the repository as a sub-directory.
This could be your home directory on your machine or a flash drive in a lab.
Windows only: To navigate in a DOS-Shell to a flash drive, you need to enter the short command:
E:

or possible another drive letter followed by a colon. DOS drive letters are annoying because they be different another
time with different resources loaded. Once you see the proper drive displayed, cd to the desired directory.
If all has gone well at bitbucket, you should be able to look at your bitbucket site and see your new repository on the
list of repositories .
For example, the co-author’s new repository, gkt170, shows up on the list of repositories (the dropdown) as
gkthiruvathukal/gkt170.
So you can now go ahead by selecting this newly created repository from the list of repositories. If all goes well, you
should see the following screen:

Somewhere on this screen, you should see this text:

Clone this repository (size: 546 bytes): HTTPS / SSH
hg clone https://[email protected]/yourusername/yourrepository

224 Chapter 18. Appendix

 Introduction to Computer Science in C#, Release 2.0

Copy the command you see in the browser starting hg clone, and paste it in as a command in your terminal/DOS-
shell window.
hg clone https://[email protected]/gkthiruvathukal/gkt170

You will see some output:

http authorization required
realm: Bitbucket.org HTTP
user: gkthiruvathukal
password:
destination directory: gkt170
no changes found
updating to branch default
0 files updated, 0 files merged, 0 files removed, 0 files unresolved

You have created a copy of the (empty) bitbucket repository in a subdirectory named the same as yourrepository
(gkt170 in the example). The is the “checkout directory”, the top level of your copy.
Again, because the repository at Bitbucket is presently empty, the above output actually makes sense. There are no
files to be updated. We’ll learn more about what this output means later. It is possible to get unresolved files when you
make changes that introduce conflicts. We’re going to do whatever we can to avoid these for the small projects in our
course work. However, when working in teams, it will become especially important that you and your teammate(s)
are careful to communicate changes you are making, especially when changing the same files in a project.

Warning: A version control system doesn’t replace the need for human communication and being organized.

Add an .hgignore and Hello World file to your project

Change directory into the top-level directory of your local repository. That should mean cd to the directory whose
name matches the bitbucket repository name.
The following is an example of a “dot hgignore” file. Mercurial will neither list or otherwise pay attention the files in
this list:
# This indicates that we are using shell-like matching logic
# instead of regular expressions.
syntax: glob
# For Mac users
Thumbs.db
.DS_Store
# This is where Xamarin Studio puts compiled stuff.
bin/
In case you compiled your own stuff, we ignore *.exe and *.dll
*.exe
*.dll
# This is a temporary debugging file generated by Xamarin Studio
*.pidb
# And one other thing we don’t need.
*.userprefs

Here is a brief explanation of what we’ve included here and why:

• syntax: glob indicates that uses the “glob” syntax, which comes from MS-DOS (the command prompt
still found on Windows). Glob syntax allows you to do special things like match all files having a certain
extension (e.g. *.exe matches Hello.exe and any other filename with extension .exe.)

18.9. Lab: Version Control 225

Introduction to Computer Science in C#, Release 2.0

• Thumbs.db and .DS_Store. Unfortunately, the Mac is still notorious for generating temporary files that
serve no purpose, except on OS X. In general, we try to keep these files out of our repository and encourage you
to do the same, especially if you are a Mac user.
• *.exe and *.dll. Anything that can be (re)produced by the Mono or Xamarin Studio tools should be ex-
cluded. In particular, do not keep these files in your repository. Today, they are quite small, but in future
development work, they can be large. Worse, they are not text files (unlike your .cs files), so they cannot be
stored optimally in a version control system.
• There are some other files produced by Xamarin Studio that we’ve put on the exclusion list, including *.pidb
and *.userprefs. The reasoning for not keeping these is similar to that in the previous case.
Now do the following steps:
1. Using your text editor, create a file .hgignore. You can simply copy and paste the above contents into this
file. Be careful of an editor like notepad, which adds ”.txt” to the end of file names by default.
Windows: To change from the default extension, use Save As, and change the file type from .txt by
electing the drop-down menu beside file type, and select “All files”.
1. Create or copy your existing Hello World example, hello.cs to the the labs folder.
2. Let’s test whether .hgignore is having any effect. Go to the labs folder and compile the Hello, World.
example.
3. Verify that the .cs and .exe files are in the labs directory (ls on Linux or OS X; dir on MS-DOS):
gkt@gkt-mini:~/gkt170/labs$ mcs hello.cs
gkt@gkt-mini:~/gkt170/labs$ ls -l
total 8
-rw-r--r-- 1 gkt gkt 224 2012-02-20 20:02 hello.cs
-rwxrwxr-x 1 gkt gkt 3072 2012-02-20 20:05 hello.exe

4. Check the status:

gkt@gkt-mini:~/gkt170/labs$ hg status
? .hgignore
? labs/hello.cs

What this tells us is that .hgignore and labs/hello.cs are not presently being tracked by our version
control system, Mercurial. The file labs/hello.exe is not shown, because it’s on the ignore list.
Note that we actually need to put the .hgignore file under version control if we want to use it wherever we
happen to be working with our stuff (i.e. when we’re not in the computer lab but, say, at home).
5. Add the file to version control:
gkt@gkt-mini:~/gkt170$ hg add .hgignore

gkt@gkt-mini:~/gkt170$ hg add labs/hello.cs

6. Commit the changes, and then see the log entry with the commands below. If you set the .hgrc file, the command
somewhere inside your local repository could be:
hg commit -m "adding an .hgignore file and Hello, World to the project"

If you did not create .hgrc, you need also include identification with -u yourName after commit, as in
hg commit -u gkt -m “adding an .hgignore file and Hello, World to the project”
It is Ok if your message wraps to a new line. You can check the log entry created by your commit:

226 Chapter 18. Appendix

 Introduction to Computer Science in C#, Release 2.0

gkt@gkt-mini:~/gkt170$ hg log
changeset: 0:9fe6ee1bf907
tag: tip
user: George K. Thiruvathukal <[email protected]>
date: Mon Feb 20 20:14:42 2012 -0600
summary: adding an .hgignore file and Hello, World to the project

7. Push the changes to Bitbucket with the following command. (You’ll be prompted for user/password not shown
here):
hg push

You should get a response like:

pushing to https://[email protected]/gkthiruvathukal/gkt170
searching for changes
remote: adding changesets
remote: adding manifests
remote: adding file changes
remote: added 1 changesets with 2 changes to 2 files
remote: bb/acl: gkthiruvathukal is allowed. accepted payload.

Create an initial structure for your project

We suggest that you follow a scheme similar to what we use when working with version control. We suggest that your
source code goes in one or more folders, like work, or homework and labs.
So let’s do it:
1. Make sure you are in the checkout directory, or cd to it.
2. Create directories:
mkdir hw
mkdir labs

We will be creating items in each one of these folders during the lab.

Warning: Please note that most version control systems do not allow you to add empty folders to the
repository. You must create at least one file and hg add it to the repository (and hg add and hg push) for the
folder to actually be created. The above was just intended to make you aware of a desired “organization”.
You are free to organize your project any way you like as long as we are able to find your homework
assignments.

Create and Test Content

1. copy in or create a simple program in a directory you created, like labs/hello.cs

2. Run it.
3. Now go back to the command prompt, and enterEnter:
hg status
and produce a response like:
? labs/hello.cs

18.9. Lab: Version Control 227

Introduction to Computer Science in C#, Release 2.0

Mercurial shows you the tracked files that are modified (none here) or files not not being tracked (after a ‘?’),
except for those files explicitly ignored. As you can see, the source (.cs) file is shown, but no “binary” objects
(like an .exe file), since you gave instructions to ignore all such files.
4. Add the new solution/projects to Mercurial. At this point, if the above list looks “reasonable” to you, you can
go ahead and just add everything. The hg command makes this easy for you. Instead of adding the specific files,
you can just type the following (nothing after the add):
hg add
and produce a response like:
adding do_the_math.cs

If you inadvertently added something that you truly don’t want in the repository, you can use the hg rm com-
mand to remove it. We have nothing at the moment that we want to remove, but want to make you aware that
correcting mistakes is possible.
5. As before, commit and push. Here is a sequence from Dr. Thiruvathukal’s Mac:
gkt@gkt-mini:~/gkt170$ hg commit -m "adding hello program"
gkt@gkt-mini:~/gkt170$ hg push
pushing to https://[email protected]/gkthiruvathukal/gkt170\
searching for changes
remote: adding changesets
remote: adding manifests
remote: adding file changes
remote: added 1 changesets with 1 change to 1 file
remote: bb/acl: gkthiruvathukal is allowed. accepted payload.

Verify that your stuff really made it to bitbucket.org

At this point, it is entirely possible that you need some convincing to believe that everything we’ve been doing thus
far is really making it to your repository at Bitbucket. Luckily, this is where having a web interface really can help us.
Do the following:
1. Log into bitbucket.org if you are not already logged in.
2. Go to Repositories and look for your repository. In the authors case, it is under gkthiruvathukal /
gkt170.
3. It pays to take a quick look at the dashboard. You’ll see the recent commits on the main screen. You should see
at least two commits from our lab session thus far, both of which likely happened just “minutes ago”.
4. You can click on any revision to see what changes were made. It is ok to do so at this time, but we’re going to
take a look at the powerful capability of “looking at the source”. So go to the Source tab.

228 Chapter 18. Appendix

 Introduction to Computer Science in C#, Release 2.0

5. If all was done properly, you will see .hgignore and labs. These were all the result of our earlier sequence of
commit+push operations. You can click on any folder to drill into the hierarchy of folders/files that have been
pushed to Bitbucket (from your local repository). In labs you find your source code (for hello.cs). Then you can
look at it–through the web! When you do so, you’ll see something like this.

18.9. Lab: Version Control 229

Introduction to Computer Science in C#, Release 2.0

Working between lab and home (or home and lab)

It may not be immediately obvious, but what we have just shown you is how to work between the classroom/lab
environment and home. In the typical scenario, when you go to your desk (or laptop), you will go through the
following lifecycle:
• hg pull: To gather any changes that you made at another location. You are always pulling changes from the
repository stored at Bitbucket, which is acting as our intermediary. Being “in the cloud” it is a great place to
keep stuff without having to worry (for the most part) about the repository getting lost.
• hg update: To update your local copy of the repository with all of the changes that you just pulled down from
Bitbucket.
• Create or modify your folders/files as desired.
• If any files that you want included were just created, use hg add. It does not hurt to use this command, even if
nothing was added.
• hg commit -m message.: Save any changes you’ve made, to your local repository only.
• hg push: Push the changes you’ve stashed in your local repository to the Bitbucket repository.
You might wonder why the pull/update and commit/push operations are separate. For a team of one (or two, if you
have a pair), it is not likely that you’d make a mistake when coordinating changes to a central repository. In a larger
team, however, some coordination is required. We’re not going into all of those details in this lab, of course, but will

230 Chapter 18. Appendix

 Introduction to Computer Science in C#, Release 2.0

likely revisit this topic as we get closer to the team project, which we think will make you thankful for having a version
control system.

Getting our book, examples, and projects

We’re going to conclude by taking this opportunity to introduce you to how we (Drs. Harrington and Thiruvathukal)
are actually using the stuff we are teaching to work as a team on developing the book and examples.
1. Pick a different location (outside of your repository folder and its subfolders) to check out our stuff from Bit-
bucket:
hg clone https://[email protected]/gkthiruvathukal/introcs-csharp

2. Don’t worry about breaking anything. Because Bitbucket knows what users are allowed to push changes to our
repository, anything you change in your copy won’t affect us. You probably are interested in how to grab our
examples and Xamarin Studio projects. If you visit our site at Bitbucket, you would see a screen like this when
viewing our repository:

3. There are several folders, but the ones of interest to you include examples and projects, where we keep
our basic examples and Xamarin Studio projects, respectively.
4. For example, if you performed a clone to introcs-csharp, you should be able to change directory to introcs-
csharp/source/examples to see all of our code examples:

18.9. Lab: Version Control 231

Introduction to Computer Science in C#, Release 2.0

gkt@gkt-mini:~/introcs-csharp/source/examples$ ls
addition1
...
write_test

(Most output has been eliminated for conciseness.)

5. You can explore introcs-csharp/source/examples to see our programs.
6. There are other folders, too. The rst folder contains the “source code” for the book itself. The devel folder
contains scripts to build the HTML, PDF, and ePub versions of our book–using a cool system named Sphinx
(from http://sphinx.pocoo.org). It’s well beyond the scope of our course to talk about this in any kind of detail
but suffice it to say, we use version control to coordinate our work to create these materials and will continue to
do so when it comes to making improvements in this and future courses.

18.10 Mercurial and Teamwork

As this course has a team project, this lecture is about how to work as a team and make effective use of the version
control system, Mercurial, that we have been using throughout the semester. While the focus is on the use of Mercurial,
the principles we are introducing here can be adapted to other situations (and alternate version control systems).

18.10.1 Planning and Communication

Two of the most important aspects when it comes to teamwork are planning and communication. In the real world,
planning is often referred to as project management. And communication often takes the form of regular team meet-
ings.
In later courses (e.g. software engineering) there is greater emphasis placed on thinking more broadly about software
process. We’re not going to cover SE in depth here but want you to be aware that software process is an important
topic. Planning and communication are always supporting ingredients of a good software process.
At the level of actual programming, when two folks are working on the same project, it is important to think about how
you can organize your work so each person on the team can get something done. As we’ve been learning throughout
the semester, the C# language gives us a way to organize our code using projects (with Xamarin Studio). Within a
project we can organize it as a collection of files, each of which maintains part of the solution to a problem. These
parts are typically organized using classes within a namespace and methods.
So the key to working together–and apart–is to spend some time, initially, planning out the essential organization of
a project and the files within it. Much like writing a term paper, you can create classes and methods that are needed–
without writing the actual body of the methods–and then commit your code to the repository. Then each member of the
team can work on parts of the code and test them independently. Then you can sit together again as a pair to integrate
the work you’ve done independently.
It’s easier said than done, but this is intended as a suggestion for how to collaborate.
In any event, the above suggests that you actually need to communicate if you want to get anything done. You should
start by discussing what needs to be done and work in real time to do what has been described above. Then you start
coding. As you are coding, you are going to realize that as well as you planned the work to be done, you “forgot” or
“misunderstood” some aspect. When this happens, you and your partner(s) need to communicate.
In the modern era of software development, we are richly blessed with synchronous communication methods such as
instant messaging, texting, group chat, and other forms of synchronous collaboration. When you and your partner(s)
are working on the project, we highly encourage you to keep a chat window open (Google Talk, AIM, Yahoo, IRC,
whatever) and use it to communicate any issues as they arise.

232 Chapter 18. Appendix

 Introduction to Computer Science in C#, Release 2.0

18.10.2 Typical Scenario

In general, when working with Mercurial, you will find yourself using the following commands in roughly this order:
• hg incoming: look for incoming changes that were either made by yourself or your partner(s). We say “by
yourself” because it is clearly possible that you are working on another computer somewhere else (laptop, home
computer) and pushed some changes. So it is a good habit to see whether “anything has changed” since the last
time you looked, even if you looked recently.
• hg pull: If there are incoming changes showing up as incoming, then you should in fact pull them in. This
will stage the changes locally but they will not be incorporated until you type hg update.
• hg update: Absorbs all changes that were pulled from your remote repository. This operation has
the potential to overwrite changes you are currently making, so you should make sure that the in-
coming changes that you observed above are sensible. For example, if you are editing a file named
examples/MyProject/my_project.cs and the incoming log suggests that the same file has been modi-
fied, you’re likely to end up with a conflict when performing the update. (We’ll cover conflict resolution shortly.)
• hg add: Whenever you add files to your folder, if you want them to be in the repository, you always need to
add them. It is very easy to forget to do so. The hg status command can be used to figure out files that
might need to be added.
• hg status: This command will become one of your best friends. You This typically tells you what has
happened since you started your work in this directory (and since the last commit/push cycle). You especially
want to keep on the lookout for the following:
– ?: This means that a file is untracked in the repository. If you see a file with the ? status that is important,
you’ll want to add it using hg add as explained above.
– M: This means that a file has been modified.
– A: This means that a file has been added.
• hg commit: In general, you should commit all changes to your repository, especially before you leave the lab
for the day. It is important to note that committing is a local operation and does not affect the remote repository
(at bitbucket.org) until a corresponding push has been done.
• hg push: Almost immediately after a commit, it makes sense to do a push, especially if you are in the lab and
will be continuing your work on other computers. In addition, as you’ve observed with previous homework, it
is the only way to ensure that you can view the code on BitBucket (our hosting site for Mercurial projects).

18.10.3 Conflict Avoidance

Inevitably, when you are working on project, you are likely to end up in a situation where you and your partner(s)
make changes that conflict with one another. In many cases, the version control software can automatically merge the
changes. Here are just a few examples of where it is possible to do so:
• You make changes to different files.
• You make additions to the repository.
• You make changes to a common file that do not overlap. An example of this might be where you have two
functions in your program and both you and your partner are careful not to modify them.
In general, we encourage you to coordinate your efforts, especially when you are doing something like the third
situation.
Where you get into trouble is when there are changes to a common file that conflict with one another. When this
happens, you have two choices in practice:
• use hg merge and hg resolve to merge your changes.

18.10. Mercurial and Teamwork 233

Introduction to Computer Science in C#, Release 2.0

• make a copy of the conflicting files (e.g. Copy hello.cs to hello.cs-backup and use hg update
--clean (changing your copy to match the current version in the remote repository) to just accept the latest
versions of all files from the repository.
In our experience, the first option is tricky. You are given the option to perform the merge anyway or use a merge tool
to select the changes of interest (and decide between them).
The second option basically results in two copies of the file. You can open up your editor to compare the files side-by-
side or use a tool like diff on Unix or a Mac, which gives a side-by-side comparison:
diff -y hello.cs my_hello.cs

(You will need to expand the width of your console window to see clearly!)
This tool is not built into Windows. In the Windows lab, you can use the much less visually helpful
fc hello.cs my_hello.cs

to show differing segments from each file. You can also download difference display tools for Windows that are more
visually helpful. One of many choices is at http://winmerge.org.

18.10.4 E-mail Notifications

One of the best ways to avoid conflicts when working on a team is to enable e-mail notification on your repository.
Bitbucket, the hosting service we are using and recommending for our students, provides full support for e-mail
notification. Whenever you or your partner(s) push changes to the hosted repository, an e-mail will be generated.
These are the steps to set it up. (Owing to the changing nature of web interfaces, we are providing generic instructions
that should be adaptable if the Bitbucket service decides to change its web user interface.)
1. Make sure your repository is selected. This is always the especially when you visit your repository by URL.
2. Select the administrative (Admin) tab.
3. Select Services (left-hand-side navigation).
4. Add the Email or Email Diff service. These services are basically equivalent, but one will generate links so you
can view the differences that were just pushed. We recommend Email Diff.
5. Add the email notification address. You can only have one address. A good way to overcome this limitation is
to set up a group service, say, at Google Groups.

18.10.5 Communication is Key to Success

At the risk of repeating ourselves, we close by reminding you of the central importance of good communication. The
authors of this book communicate when it comes to their changes–even before we make them. Yet we occasionally
trip over each other, and there is usually a fair amount of manual reconciliation required to deal with conflicts when
we end up touching the same file by mistake.
When you absolutely and positively need to change a common file, it is important to ask yourself the important
question: Shouldn’t we be sitting together to make these changes? It’s a rhetorical question, but working closely
together, either in the same room or through a chat session/phone call, can result in significantly fewer headaches,
especially during the early stages of a project.
So please take this time to stop what you are doing and communicate. You’ll know your communication is good if
you never need to do anything that has been described on this page. Then again, we’re human. So you it is likely to
happen at least once. (We know from experience but are doing everything possible to avoid conflicts in our work!)

234 Chapter 18. Appendix

 Introduction to Computer Science in C#, Release 2.0

18.11 Recent/Current Course Offerings

18.11.1 Dr. Harrington’s Offerings

• Dr. Harrington’s COMP 170 Site

18.11.2 Dr. Thiruvathukal’s Offerings

• Fall 2012 Syllabus

• Fall 2012 Schedule

18.12 Acknowledgments

• The Sphinx team at http://sphinx.pocoo.org for creating such an awesome documentation tool
• The Bitbucket team at http://bitbucket.org for giving us a great place to collaborate.
• Jeremy Chalmer at http://jchalmer.com/ for his work to get a nice Twitter Bootstrap theme working with Sphinx.
• ...we’ll of course acknowledge our significant others/families when we’re done.

18.11. Recent/Current Course Offerings 235

Introduction to Computer Science in C#, Release 2.0

236 Chapter 18. Appendix

 BIBLIOGRAPHY

[WirthADP] Niklaus Wirth, Algorithms + Data Structures = Programs, Prentice Hall, 1976.
[WikipediaShellSort] http://en.wikipedia.org/wiki/Shellsort
[uClibc] http://en.wikipedia.org/wiki/UClibc
[TCSortingJava] http://tools-of-computing.com/tc/CS/Sorts/SortAlgorithms.htm
[CamelCase] http://en.wikipedia.org/wiki/CamelCase
[UPennCSharp] http://www.cis.upenn.edu/~cis193/csstyle.html

237
Introduction to Computer Science in C#, Release 2.0

238 Bibliography
 INDEX

Symbols < less than, 63

() <= less than or equal, 63
function call, 52 {}
function definition, 51 compound statement, 61
grouping, 16 Format, 29
* multiplication, 15 scope, 62
*/ end comment, 38 []
*= operator, 106 array, 127
+ attribute, 181
arithmetic, 15 dictionary, 159
string, 23
+= operator, 106 A
- abstraction, 44
negation, 16 actual
subtraction, 15 parameter, 45
-= operator, 106 addition2.cs
. example, 47
class static member, 24 Agree
double literal, 16 exercise, 86
object’s method, 118 algorithms
part of namespace, 117 array, 136
.net api, 158 binary search, 144
/ division, 16 bubble sort, 137
/* ... */ comment, 38 greatest common divisor, 91
// comment, 7, 38 insertion sort, 139
/= operator, 106 linear search, 136
= Quicksort, 141
assignment statement, 18 selection sort, 138
initializer, 18 alias, 131
== equality test, 63 and &&, 69
% remainder, 16 arithmetic, 15
%= operator, 106 array, 125
&& [ ], 127
boolean operation AND, 69 algorithms, 136
short-circuit, 87 anonymous initialization, 132
\ escape code, 28 Dups, 133
|| Histogram, 134
boolean operation OR, 70 Mirror, 133
short-circuit, 87 nested loop, 137–141
> greater than, 63 of arrays, 152
>= greater than or equal, 63 one dimensional, 127
< > for generics, 155 Reverse, 134

239
Introduction to Computer Science in C#, Release 2.0

TrimAll, 133 history, 206

two dimensional array, 150 parameters, 129
two-dimensional, ragged, 152 parameters in Xamarin Studio, 129
array parameter paths, 203
Scale, 131 shortcuts, 206
ASCII example, 109 command line adder
assertion example, 133
testing, 181 comment, 7, 38
assignment statement, 22 comparison
=, 18 < > <= >= ==
attribute =, 63
[ ], 181 compile
command line, 123
B compiler error
big oh uninitialized local variable, 63
constant order, 78, 160 compound statement
dictionary, 160 { }, 61
linear order, 160 scope, 62
order of n, 78 concatenation
binary string, 23
search, 144 concrete example, 76
bitbucket.org, 221 splitting a loop, 84
book alternate formats, 3 Console
book examples download, 2 ReadKey, 26
booklist Write, 24
homework, 217 WriteLine, 24
boolean expression constant
comparison, 63 global, 48
boolean operation constant order
&& AND, 69 big oh, 78, 160
|| OR, 70 constructor
break statement, 104 OOP, 165, 167
bubble sort consumer
algorithms, 137 function, 47
sorting, 137 Contains
string, 81
C ContainsKey
camel case dictionary, 160
identifier, 21 continue statement, 104
naming convention, 21 copy file to upper case
case sensitive, 21 example, 120
cast, 32 CountRep
pitfall, 79 exercise, 90
char csharp, 15, 19
type, 32 help, 19
underlying numeric code, 102 mono command prompt (Windows), 201
class, 164 quit, 19
Rational, 165 ShowVars, 19
StreamReader, 118
StreamWriter, 117
D
command line, 202 dangling else;
compile, 123 pitfall, 68
execution, 122 decision
file completion, 206 homework, 206, 210

240 Index
 Introduction to Computer Science in C#, Release 2.0

declaration statement, 22 ReadLines, 157

default value Reverse, 134
OOP, 132 Scale, 131
definition sum_files.cs, 119
function, 41 TrimAll, 133
development tools, 199 Word Count, 161
dictionary, 158 examples download, 2
[ ], 159 execution
big oh, 160 command line, 122
ContainsKey, 160 execution sequence
Keys, 160 function, 42, 45
Directory, 122 exercise
division Agree, 86
pitfall, 78 CountRep, 90
division /, 16 GroupFlips, 113
do while heads or tails, 113
statement, 94 row and column numbering, 152
documentation safe PromptDouble, 90
function, 50 safe PromptInt, 90
double, 16 two dimensional array, 152
Parse, 26 varying column width, 152
type, 31 expression, 17
double example
IsDigits, 81 F
double this Factorial, 114
OOP, 168 field
Dups private, 166
array, 133 field width formatting, 107
example, 133 File, 122
file, 115
E abstraction, 117
edge case close, 117, 118
range testing, 78 read, 118
testing, 78 ReadToEnd, 119
EndOfStream stream, 117
StreamReader, 119 write, 117
EndsWith file completion
string, 90 command line, 206
escape code \, 28 FIO
Euclid file I/O, 124
history, 91 for
example loop, 102
addition2.cs, 47 foreach
ASCII, 109 loop, 99
command line adder, 133 statement, 101
copy file to upper case, 120 formal
Dups, 133 parameter, 45
HashSet, 161 Format
Histogram, 134 { }, 29
List, 157, 161 string, 86
Mirror, 133 format
mod_mult_table.cs, 109 field width, 107
OneCharPerLine, 80 left justification, 108
power_table.cs, 107 literal {}, 30

Index 241
Introduction to Computer Science in C#, Release 2.0

table, 109 input-output, 206, 210

function loop, 210
consumer, 47 OOP, 217
definition, 41
documentation, 50 I
execution sequence, 42, 45 IComparable Interface, 185
not use return value, 48 identifier, 21
parameter, 43, 45 camel case, 21
return, 45 multiple word, 21
scope, 48 naming convention, 21
summary, 51 underscores, 21
writer, 47 if
function syntax, 51 need braces, 69
pitfall, 67
G statement, 59
Gauss if-else, 62
sum through n, 78 dangling else pitfall, 68
generic if-else-if, 65
HashSet, 161 immutable, 57
generics, 155 index
getter sequence, 79
OOP, 168 string, 55
global while, 79
constant, 48 IndexOf
grade files string, 89
homework, 213 infinite loop
greatest common divisor pitfall, 86
algorithms, 91 initializer
iterative, 94 =, 18
recursion, 94 input-output
grouping homework, 206, 210
( ), 16 insertion sort
algorithms, 139
H sorting, 139
HashSet installation
example, 161 mono, 200
generic, 161 instance method
set, 161 OOP, 168
heads or tails this, 168
exercise, 113 instance variable
random, 113 private, 166
hg int, 16
labs, 220 Parse, 26
Histogram value range, 31
array, 134 interactive
example, 134 loop, 83
history while, 83
command line, 206 interface
Euclid, 91 IComparable, 185
SP1, 91, 136 IntroCS
homework namespace, 50
booklist, 217 introduction, 5
decision, 206, 210 IntsFromString, 130
grade files, 213

242 Index
 Introduction to Computer Science in C#, Release 2.0

K Mercurial, 232
Keys method
dictionary, 160 overloading, 30
keyword, 21 string, 55
Mirror
L array, 133
labs example, 133
arrays, 145 mod_mult_table.cs
division sentences, 37 example, 109
hg, 220 mono
loops, 113 installation, 200
string manipulations, 59, 95 mono command prompt (Windows), 201
version control, 220 csharp, 201
Xamarin Studio, 8 multiple source files
left justification library class, 49
format, 108 multiple word
library identifier, 21
UI, 91 underscores, 21
UIF, 49
library class
N
multiple source files, 49 namespace
limit on number size IntroCS, 50
pitfall, 78 naming convention
linear order camel case, 21
big oh, 160 identifier, 21
linear search NAnt build tool, 124
algorithms, 136 need braces
List if, 69
Add, 155 nested loop
constructor, 155 table, 109
constructor with sequence, 157 not , 70
Contains, 155 not use return value
Count, 155 function, 48
example, 157, 161 null from ReadLine
Remove, 155 StreamReader, 119
RemoveAt, 155 numeric type
list, 155 range, 32
literal, 21
local O
scope, 47 OneCharPerLine
long example, 80
type, 31 OOP
loop constructor, 165, 167
for, 102 default value, 132
foreach, 99 field, instance variable, 166
homework, 210 getter, 168
interactive, 83 homework, 217
nested, 114 instance method, 168
while, 71, 83 private helping function, 168
operator
M *, 15
Main + (with numbers), 15
parameters, 129, 133 += -= *= /= %=, 106
mcs, 123 -, 15

Index 243
Introduction to Computer Science in C#, Release 2.0

/, %, 16 program structure, 24
or ||, 70 public, 50
order of n
big oh, 78 Q
overloading questions
method, 30 while, 75
override, 169 Quicksort
algorithms, 141
P sorting, 141
parameter
actual, 45 R
formal, 45 Random, 141
function, 43, 45 random
parameters heads or tails, 113
command line, 129 random number
Main, 129, 133 generator, 97
Parse seed, 141
double, 26 range
int, 26 numeric type, 32
path, 121 range testing
paths edge case, 78
command line, 203 testing, 78
performance Rational
Stopwatch, 142 class, 165
TimeSpan, 142 read
PF4 file, 118
recursion, 91, 136 ReadKey, 26
pitfall ReadLine
cast, 79 StreamReader, 118
dangling else;, 68 ReadLines
division, 78 example, 157
if, 67 ReadToEnd
infinite loop, 86 file, 119
limit on number size, 78 StreamReader, 119
need braces for if, 69 recursion, 141
repeat declaration, 85 greatest common divisor, 94
repeat interactive input, 85 PF4, 91, 136
power_table.cs reference object
example, 107 value object, 175
precedence, 206 regenerate random numbers, 141
precision, 107 remainder %, 16
formatting, 107 repeat declaration
PrintRectangle, 114 pitfall, 85
PrintReps, 113 repeat interactive input
PrintVowels pitfall, 85
string, 80 Replace
private string, 90
field, 166 return
instance variable, 166 from inside loop, 83
private helping function function, 45
OOP, 168 Reverse
problem solving array, 134
strategy, 35 example, 134
string, 57 row and column numbering

244 Index
 Introduction to Computer Science in C#, Release 2.0

exercise, 152 StartsWith

two dimensional array, 152 string, 90
rubric statement
while, 74 assignment, 18, 22
break, 104
S compound, 61
safe PromptDouble continue, 104
exercise, 90 declaration, 22
safe PromptInt do while, 94
exercise, 90 for, 103
Scale foreach, 101
array parameter, 131 if, 59
example, 131 while, 73
scope Stopwatch
compound statement, 62 performance, 142
function, 48 timing, 142
local, 47 stream
search file, 117
binary, 144 StreamReader
linear, 136 class, 118
seed EndOfStream, 119
random number, 141 null from ReadLine, 119
selection sort ReadLine, 118
algorithms, 138 ReadToEnd, 119
sorting, 138 StreamWriter
semicolon after condition class, 117
pitfall, 67 WriteLine, 117
sequence string, 28
index, 79 +, 23
while, 79 |hyperpage, 23
set concatenation, 23
HashSet, 161 Contains, 81
Shell sort convert to double, 26
sorting, 140 convert to int, 26
short-circuit EndsWith, 90
&&, 87 Format, 86
||, 87 index, 55
shortcuts IndexOf, 89
command line, 206 method, 55, 89
side effect, 157 PrintVowels, 80
sorting problem solving, 57
bubble sort, 137 Replace, 90
insertion sort, 139 Split, 130
Quicksort, 141 StartsWith, 90
selection sort, 138 Trim, 89
Shell sort, 140 StringOfReps, 114
source download, 2 struct
source.zip, 2 value object, 175
SP1 sum through n
history, 91, 136 Gauss, 78
Split sum_files.cs
string, 130 example, 119
splitting a loop summary
concrete example, 84 function, 51

Index 245
Introduction to Computer Science in C#, Release 2.0

syntax template var

typography, 22 type, 119
variable
T assignment, =, 18
table varying column width
format, 107, 109 exercise, 152
nested loop, 109 two dimensional array, 152
tables, 107 version control, 232
testing, 179 labs, 220
assertion, 181
edge case, 78 W
range testing, 78 warning
this redeclaring instance variables, 174
instance method, 168 while
TimeSpan do while, 94
performance, 142 index, 79
timing interactive, 83
Stopwatch, 142 loop, 71
ToString, 169 questions, 75
Trim rubric, 74
string, 89 sequence, 79
TrimAll statement, 73
array, 133 whitespace, 7
example, 133 Word Count
two dimensional array example, 161
array, 150 Write
exercise, 152 Console, 24
row and column numbering, 152 write
varying column width, 152 file, 117
type WriteLine
char, 32 { }, 29
double, 16, 31 Console, 24
int, 16 StreamWriter, 117
long, 31 writer
value, 30 function, 47
var, 119
typography X
syntax template, 22 Xamarin Studio, 8
command line parameters, 129
U empty project - no console error, 39
UI file not in project error, 39
library, 91 further tools, 202
underscores installation, 200
identifier, 21
multiple word, 21

V
value
type, 30
value object
reference object, 175
struct, 175
value range
int, 31

246 Index