0% found this document useful (0 votes)

10K views

Clean Code in Csharp PDF

Uploaded by

Rodrigo García

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

10K views

Clean Code in Csharp PDF

Uploaded by

Rodrigo García

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 487

Clean Code in C#

Refactor your legacy C# code base and improve application

performance by applying best practices

Jason Alls

BIRMINGHAM - MUMBAI
Clean Code in C#
Copyright © 2020 Packt Publishing

All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form
or by any means, without the prior written permission of the publisher, except in the case of brief quotations
embedded in critical articles or reviews.

Every effort has been made in the preparation of this book to ensure the accuracy of the information presented.
However, the information contained in this book is sold without warranty, either express or implied. Neither the
author, nor Packt Publishing or its dealers and distributors, will be held liable for any damages caused or alleged to
have been caused directly or indirectly by this book.

Packt Publishing has endeavored to provide trademark information about all of the companies and products
mentioned in this book by the appropriate use of capitals. However, Packt Publishing cannot guarantee the accuracy
of this information.

Commissioning Editor: Richa Tripathi

Acquisition Editor: Alok Dhuri
Content Development Editor: Ruvika Rao
Senior Editor: Nitee Shetty
Technical Editor: Pradeep Sahu
Copy Editor: Safis Editing
Project Coordinator: Francy Puthiry
Proofreader: Safis Editing
Indexer: Manju Arasan
Production Designer: Nilesh Mohite

First published: July 2020

Production reference: 1170720

Published by Packt Publishing Ltd.

Livery Place
35 Livery Street
Birmingham
B3 2PB, UK.

ISBN 978-1-83898-297-3

www.packt.com
To my parents, for supporting me throughout my life and career. To all the people in the world
of software that have made my career possible, and who have employed me, trained me, and
worked alongside me. You have been instrumental in helping me to get to where I am today.
I thank you all.
Packt.com

Subscribe to our online digital library for full access to over 7,000 books and videos, as well
as industry leading tools to help you plan your personal development and advance your
career. For more information, please visit our website.

Why subscribe?
Spend less time learning and more time coding with practical eBooks and Videos
from over 4,000 industry professionals

Improve your learning with Skill Plans built especially for you

Get a free eBook or video every month

Fully searchable for easy access to vital information

Copy and paste, print, and bookmark content

Did you know that Packt offers eBook versions of every book published, with PDF and
ePub files available? You can upgrade to the eBook version at www.packt.com and as a print
book customer, you are entitled to a discount on the eBook copy. Get in touch with us at
[email protected] for more details.

At www.packt.com, you can also read a collection of free technical articles, sign up for a
range of free newsletters, and receive exclusive discounts and offers on Packt books and
eBooks.
Contributors

About the author

Jason Alls has been programming for over 21 years using Microsoft technologies. Working
with an Australasian company, he started his career developing call center management
reporting software used by global clients including telecom providers, banks, airlines, and
the police. He then moved on to develop GIS marketing applications and worked in the
banking sector performing data migrations between Oracle and SQL Server. Certified as an
MCAD in C# since 2005, he has been involved in the development of various desktop, web,
and mobile applications.

Currently employed by a globally recognized leader in the educational software sector, he

develops and supports dyslexia testing and assessment software written in ASP.NET,
Angular, and C#.

I would like to thank my parents for always being there, supporting me throughout my life
and career. Career-wise, I would like to thank all the people in the world of computing that
have made my career possible. Especially those who have employed me, trained me, and
worked alongside me. You have helped me to get to where I am today.

A special thank you to all the staff at Packt Publishing who provided me with the
opportunity to write this book, and who assisted me in improving the content. It has been
an eye-opening experience and a pleasant one. It is your hard work and dedication to the
book-writing process that enables computer programmers like me to become accomplished
authors. This book would not be what it is without your valuable input.
About the reviewer
Omprakash Pandey, a Microsoft 365 consultant, has been working with industry experts to
understand project requirements and work on the implementation of projects for the last 20
years. He has trained more than 50,000 aspiring developers and has assisted in the
development of more than 50 enterprise applications. He has offered innovative solutions
on .NET development, Microsoft Azure, and other technologies. He has worked for
multiple clients across various locations, including Hexaware, Accenture, Infosys, and
many more. He has been a Microsoft Certified Trainer for more than 5 years.

I want to thank my parents, my colleagues, Ashish, and Francy, for their assistance and
support.

Packt is searching for authors like you

If you're interested in becoming an author for Packt, please visit authors.packtpub.com
and apply today. We have worked with thousands of developers and tech professionals,
just like you, to help them share their insight with the global tech community. You can
make a general application, apply for a specific hot topic that we are recruiting an author
for, or submit your own idea.
Table of Contents
Preface 1
Chapter 1: Coding Standards and Principles in C# 7
Technical requirements 8
Good code versus bad code 9
Bad code 9
Improper indentation 10
Comments that state the obvious 10
Comments that excuse bad code 11
Commented-out lines of code 12
Improper organization of namespaces 12
Bad naming conventions 13
Classes that do multiple jobs 13
Methods that do many things 16
Methods with more than 10 lines of code 17
Methods with more than two parameters 18
Using exceptions to control program flow 18
Code that is difficult to read 19
Code that is tightly coupled 20
Low cohesion 21
Objects left hanging around 21
Use of the Finalize() method 22
Over-engineering 22
Learn to Keep It Simple, Stupid 22
Lack of regions in large classes 23
Lost-intention code 23
Directly exposing information 24
Good code 24
Proper indentation 25
Meaningful comments 25
API documentation comments 25
Proper organization using namespaces 26
Good naming conventions 26
Classes that only do one job 27
Methods that do one thing 27
Methods with less than 10 lines, and preferably no more than 4 27
Methods with no more than two parameters 28
Proper use of exceptions 28
Code that is readable 28
Code that is loosely coupled 29
High cohesion 29
Objects are cleanly disposed of 29
Avoiding the Finalize() method 30
The right level of abstraction 30
Table of Contents

Using regions in large classes 31

The need for coding standards, principles, and methodologies 31
Coding standards 31
Coding principles 32
Coding methodologies 32
Coding conventions 33
Modularity 33
KISS 34
YAGNI 34
DRY 35
SOLID 35
Occam's Razor 36
Summary 36
Questions 37
Further reading 37
Chapter 2: Code Review – Process and Importance 38
The code review process 39
Preparing code for review 40
Leading a code review 41
Issuing a pull request 43
Responding to a pull request 46
Effects of feedback on reviewees 48
Knowing what to review 51
Company's coding guidelines and business requirement(s) 51
Naming conventions 51
Formatting 52
Testing 52
Architectural guidelines and design patterns 54
Performance and security 54
Knowing when to send code for review 55
Providing and responding to review feedback 57
Providing feedback as a reviewer 57
Responding to feedback as a reviewee 58
Summary 59
Questions 59
Further reading 60
Chapter 3: Classes, Objects, and Data Structures 61
Technical requirements 62
Organizing classes 62
A class should have only one responsibility 65
Commenting for documentation generation 67
Cohesion and coupling 70
An example of tight coupling 71

[ ii ]
Table of Contents

An example of low coupling 72

An example of low cohesion 73
An example of high cohesion 74
Design for change 75
Interface-oriented programming 76
Dependency injection and inversion of control 79
An example of DI 79
An example of IoC 81
The Law of Demeter 83
A good and a bad example (chaining) of the Law of Demeter 83
Immutable objects and data structures 85
An example of an immutable type 86
Objects should hide data and expose methods 87
An example of encapsulation 87
Data structures should expose data and have no methods 88
An example of data structure 89
Summary 89
Questions 90
Further reading 90
Chapter 4: Writing Clean Functions 91
Understanding functional programming 92
Keeping methods small 95
Indenting code 97
Avoiding duplication 99
Avoiding multiple parameters 100
Implementing SRP 102
Summary 107
Questions 107
Further reading 108
Chapter 5: Exception Handling 109
Checked and unchecked exceptions 110
Avoiding NullPointerExceptions 114
Business rule exceptions 117
Example 1 – handling conditions with business rule exceptions 120
Example 2 – handling conditions with normal program flow 121
Exceptions should provide meaningful information 123
Building your own custom exceptions 125
Summary 128
Questions 129
Further reading 129
Chapter 6: Unit Testing 130
Technical Requirements 131

[ iii ]
Table of Contents

Understanding the reasons for a good test 131

Understanding the testing tools 137
MSTest 137
NUnit 145
Moq 152
SpecFlow 157
TDD methodology practice – fail, pass, and refactor 162
Removing redundant tests, comments, and dead code 169
Summary 171
Questions 171
Further reading 172
Chapter 7: End-to-End System Testing 173
E2E testing 173
The login module (subsystem) 175
The admin module (subsystem) 179
The test module (subsystem) 181
Testing our three-module system using E2E 182
Factories 185
Dependency injection 194
Modularization 200
Summary 202
Questions 203
Further reading 203
Chapter 8: Threading and Concurrency 204
Understanding the thread life cycle 205
Adding thread parameters 207
Using a thread pool 208
Task Parallel Library 209
Parallel.Invoke() 209
Parallel.For() 210
ThreadPool.QueueUserWorkItem() 212
Using a mutex with synchronous threads 212
Working with parallel threads using semaphores 215
Limiting the number of processors and threads in the thread pool 217
Preventing deadlocks 219
Coding a deadlock example 220
Preventing race conditions 225
Understanding static constructors and methods 228
Adding static constructors to our sample code 229
Adding static methods to our sample code 230
Mutability, immutability, and thread safety 234
Writing code that is mutable and not thread-safe 234
Writing code that is immutable and thread-safe 236

[ iv ]
Table of Contents

Understanding thread safety 238

Synchronized method dependencies 242
Using the Interlocked class 243
General recommendations 247
Summary 248
Questions 249
Further reading 249
Chapter 9: Designing and Developing APIs 251
Technical requirements 252
What is an API? 252
API proxies 254
API design guidelines 256
Well-defined software boundaries 259
Understanding the importance of good quality API documentation 261
Swagger API development 262
Passing immutable structs instead of mutable objects 265
Testing third-party APIs 268
Testing your own APIs 269
API design using RAML 271
Installing Atom and API Workbench by MuleSoft 272
Creating the project 274
Generating our C# API from our agnostic RAML design specification 277
Summary 281
Questions 281
Further reading 282
Chapter 10: Securing APIs with API Keys and Azure Key Vault 283
Technical requirements 284
Undertaking the API project – dividend calendar 284
Accessing the Morningstar API 286
Storing the Morningstar API key in Azure Key Vault 286
Creating the dividend calendar ASP.NET Core web application in
Azure 289
Publishing our web application 291
Using an API key to secure our dividend calendar API 297
Setting up the repository 297
Setting up authentication and authorization 300
Adding authentication 300
Adding authorization 304
Testing our API key security 308
Adding the dividend calendar code 311
Throttling our API 319
Summary 323
Questions 324

[v]
Table of Contents

Data clump 391

Deodorant comments 391
Duplicate code 392
Lost intent 392
The mutation of variables 393
The oddball solution 395
Shotgun surgery 397
Solution sprawl 399
Uncontrolled side effects 399
Class-level code smells 400
Cyclomatic complexity 400
Replacing switch statements with the factory pattern 400
Improving the readability of conditional checks within an if statement 403
Divergent change 404
Downcasting 405
Excessive literal use 405
Feature envy 405
Inappropriate intimacy 407
Indecent exposure 408
The large class (aka the God object) 408
The lazy class (aka the freeloader and the lazy object) 408
The middleman class 409
The orphan class of variables and constants 409
Primitive obsession 409
Refused bequest 410
Speculative generality 410
Tell, Don't Ask 410
Temporary fields 410
Method-level smells 411
The black sheep method 411
Cyclomatic complexity 411
Contrived complexity 411
Dead code 411
Excessive data return 412
Feature envy 412
Identifier size 412
Inappropriate intimacy 412
Long lines (aka God lines) 413
Lazy methods 413
Long methods (aka God methods) 413
Long parameter lists (aka too many parameters) 413
Message chains 413
The middleman method 414
Oddball solutions 414
Speculative generality 414
Summary 414

[ vii ]
Table of Contents

Questions 415
Further reading 416
Chapter 14: Refactoring C# Code – Implementing Design Patterns 417
Technical requirements 418
Implementing creational design patterns 418
Implementing the singleton pattern 419
Implementing the factory method pattern 420
Implementing the abstract factory pattern 422
Implementing the prototype pattern 425
Implementing the builder pattern 427
Implementing structural design patterns 433
Implementing the bridge pattern 434
Implementing the composite pattern 436
Implementing the façade pattern 439
Implementing the flyweight pattern 442
Overview of behavioral design patterns 445
Final thoughts 446
Summary 448
Questions 449
Further reading 450
Appendix A: Assessments 451
Chapter 1 451
Chapter 2 451
Chapter 3 452
Chapter 4 452
Chapter 5 453
Chapter 6 454
Chapter 7 455
Chapter 8 455
Chapter 9 456
Chapter 10 457
Chapter 11 457
Chapter 12 458
Chapter 13 458
Chapter 14 460
Other Books You May Enjoy 462
Index 465

[ viii ]
Preface
Welcome to Clean Code in C#. You will learn how to identify problematic code that, while it
compiles, does not lend itself to readability, maintainability, and extensibility. You will also
learn about various tools and patterns, along with ways to refactor code to make it clean.

Who this book is for

This book is aimed at computer programmers with a good grasp of the C# programming
language who would like guidance on identifying problematic code and writing clean code
in C#. Primarily, the reader base will range from graduate to mid-level programmers, but
even senior programmers may find this book valuable.

What this book covers

Chapter 1, Coding Standards and Principles in C#, looks at some good code contrasted with
bad code. As you read through this chapter, you will come to understand why you need
coding standards, principles, methodologies, and code conventions. You will learn about
modularity and the design guidelines KISS, YAGNI, DRY, SOLID, and Occam's razor.

Chapter 2, Code Review – Process and Importance, takes you through the code review process
and provides reasons for its importance. In this chapter, you are guided through the
process of preparing code for review, leading a code review, knowing what to review,
knowing when to send code for review, and how to provide and respond to review
feedback.

Chapter 3, Classes, Objects, and Data Structures, covers the broad topics of class
organization, documentation comments, cohesion, coupling, the Law of Demeter, and
immutable objects and data structures. By the end of the chapter, you will be able to write
code that is well organized and only has a single responsibility, provide users of the code
with relevant documentation, and make code extensible.

Chapter 4, Writing Clean Functions, helps you to understand functional programming, how
to keep methods small, and how to avoid code duplication and multiple parameters. By the
time you finish this chapter, you will be able to describe functional programming, write
functional code, avoid writing code with more than two parameters, write immutable data
objects and structures, keep your methods small, and write code that adheres to the Single
Responsibility Principle.
Preface

Chapter 5, Exception Handling, covers checked and unchecked exceptions,

NullPointerException, and how to avoid them as well as covering, business rule exceptions,
providing meaningful data, and building your own custom exceptions.

Chapter 6, Unit Testing, takes you through using the Behavior-Driven Development
(BDD) software methodology using SpecFlow, and Test-Driven Development (TDD)
using MSTest and NUnit. You will learn how to write mock (fake) objects using Moq, and
how to use the TDD software methodology to write tests that fail, make the tests pass, and
then refactor the code once it passes.

Chapter 7, End-to-End System Testing, guides you through the manual process of end-to-
end testing using an example project. In this chapter, you will perform End-to-End (E2E)
testing, code and test factories, code and test dependency injection, and test
modularization. You will also learn how to utilize modularization.

Chapter 8, Threading and Concurrency, focuses on understanding the thread life cycle;
adding parameters to threads; using ThreadPool, mutexes, and synchronous threads;
working with parallel threads using semaphores; limiting the number of threads and
processors used by ThreadPool; preventing deadlocks and race conditions; static methods
and constructors; mutability and immutability; and thread-safety.

Chapter 9, Designing and Developing APIs, helps you to understand what an API is, API
proxies, API design guidelines, API design using RAML, and Swagger API development.
In this chapter, you will design a language-agnostic API in RAML and develop it in C#, and
you will document your API using Swagger.

Chapter 10, Securing APIs with API Keys and Azure Key Vault, shows you how to obtain a
third-party API key, store that key in Azure Key Vault, and retrieve it via an API that you
will build and deploy to Azure. You will then implement API key authentication and
authorization to secure your own API.

Chapter 11, Addressing Cross-Cutting Concerns, introduces you to using PostSharp to

address cross-cutting concerns using aspects and attributes that form the basis of aspect-
oriented development. You will also learn how to use proxies and decorators.

Chapter 12, Using Tools to Improve Code Quality, exposes you to various tools that will assist
you in writing quality code and improving the quality of existing code. You'll gain
exposure to code metrics and code analysis, quick actions, the JetBrains tools called
dotTrace Profiler and Resharper, and Telerik JustDecompile.

[2]
Preface

Chapter 13, Refactoring C# Code – Identifying Code Smells, is the first of two chapters that
take you through different types of problematic code and show you how to modify it to be
clean code that is easy to read, maintain, and extend. Code problems are listed
alphabetically through each chapter. Here, you will cover such topics as class
dependencies, code that can't be modified, collections, and combinatorial explosion.

Chapter 14, Refactoring C# Code – Implementing Design Patterns, takes you through the
implementation of creational and structural design patterns. Here, behavioral design
patterns are briefly touched upon. You are then given some final thoughts on clean code
and refactoring.

To get the most out of this book

The majority of the chapters can be read independently of each other and in any order. But
to get the most out of this book, I recommend that you read the chapters in the order
presented. As you work through the chapters, follow the instructions, and carry out the
tasks presented. Then, when you reach the end of a chapter, answer the questions and carry
out the recommended further reading to reinforce what you have learned. For maximum
benefit when working through the contents of this book, it is recommended that you meet
the following requirements:

Software/hardware covered in
Requirements
the book

Visual Studio 2019 Windows 10, macOS

Atom Windows 10, macOS, Linux: https://atom.io/

Azure resources Azure subscription: https://azure.microsoft.com/en-gb/

Azure Key Vault Azure subscription: https://azure.microsoft.com/en-gb/

Obtain your own API key from https://rapidapi.com/

The Morningstar API
integraatio/api/morningstar1

Postman Windows 10, macOS, Linux: https://www.postman.com/

It will be useful if you have these in place before you start reading and working your way
through the chapters.

[3]
Preface

If you are using the digital version of this book, we advise you to type the code yourself
or access the code via the GitHub repository (link available in the next section). Doing so
will help you avoid any potential errors related to the copying and pasting of code.

You should have basic experience of using Visual Studio 2019 Community Edition or
higher, and basic C# programming skills, including writing console applications. Many
examples will be in the form of C# console applications. The main project will be using
ASP.NET. It will help if you are capable of writing ASP.NET websites using the framework
and core. But don't worry – you will be guided through the steps that you need to go
through.

Download the example code files

You can download the example code files for this book from your account
at www.packt.com. If you purchased this book elsewhere, you can
visit www.packtpub.com/support and register to have the files emailed directly to you.

You can download the code files by following these steps:

1. Log in or register at www.packt.com.

2. Select the Support tab.
3. Click on Code Downloads.
4. Enter the name of the book in the Search box and follow the onscreen
instructions.

Once the file is downloaded, please make sure that you unzip or extract the folder using the
latest version of:

WinRAR/7-Zip for Windows

Zipeg/iZip/UnRarX for Mac
7-Zip/PeaZip for Linux

The code bundle for the book is also hosted on GitHub at https://github.com/
PacktPublishing/Clean-Code-in-C-. In case there's an update to the code, it will be
updated on the existing GitHub repository.

We also have other code bundles from our rich catalog of books and videos available
at https://github.com/PacktPublishing/. Check them out!

[4]
Preface

Download the color images

We also provide a PDF file that has color images of the screenshots/diagrams used in this
book. You can download it here: https://static.packt-cdn.com/downloads/
9781838982973_ColorImages.pdf.

Conventions used
There are a number of text conventions used throughout this book.

CodeInText: Indicates code words in text, database table names, folder names, filenames,
file extensions, pathnames, dummy URLs, user input, and Twitter handles. Here is an
example: "The InMemoryRepository class implements the GetApiKey() method
of IRepository. This returns a dictionary of API keys. These keys will be stored in
our _apiKeys dictionary member variable."

A block of code is set as follows:

using CH10_DividendCalendar.Security.Authentication;
using System.Threading.Tasks;

namespace CH10_DividendCalendar.Repository
{
public interface IRepository
{
Task<ApiKey> GetApiKey(string providedApiKey);
}
}

Any command-line input or output is written as follows:

az group create --name "<YourResourceGroupName>" --location "East US"

Bold: Indicates a new term, an important word, or words that you see onscreen. For
example, words in menus or dialog boxes appear in the text like this. Here is an example:
"To create the app service, right-click the project you created and select Publish from the
menu."

[5]
Preface

Warnings or important notes appear like this.

Tips and tricks appear like this.

Get in touch
Feedback from our readers is always welcome.

General feedback: If you have questions about any aspect of this book, mention the book
title in the subject of your message and email us at [email protected].

Errata: Although we have taken every care to ensure the accuracy of our content, mistakes
do happen. If you have found a mistake in this book, we would be grateful if you would
report this to us. Please visit www.packtpub.com/support/errata, selecting your book,
clicking on the Errata Submission Form link, and entering the details.

Piracy: If you come across any illegal copies of our works in any form on the Internet, we
would be grateful if you would provide us with the location address or website name.
Please contact us at [email protected] with a link to the material.

If you are interested in becoming an author: If there is a topic that you have expertise in
and you are interested in either writing or contributing to a book, please
visit authors.packtpub.com.

Reviews
Please leave a review. Once you have read and used this book, why not leave a review on
the site that you purchased it from? Potential readers can then see and use your unbiased
opinion to make purchase decisions, we at Packt can understand what you think about our
products, and our authors can see your feedback on their book. Thank you!

For more information about Packt, please visit packt.com.

[6]
1
Coding Standards and
Principles in C#
The primary goal of coding standards and principles in C# is for programmers to become
better at their craft by programming code that is more performant and easier to maintain.
In this chapter, we will look at some examples of good code contrasted with examples of
bad code. This will lead nicely into discussing why we need coding standards, principles,
and methodologies. We will then move on to consider conventions for naming,
commenting, and formatting source code, including classes, methods, and variables.

A big program can be rather unwieldy to understand and maintain. For junior
programmers, getting to know the code and what it does can be a daunting prospect.
Teams can find it hard to work together on such projects. And from a testing viewpoint, it
can make things rather difficult. Because of this, we will look at how you use modularity to
break programs down into smaller modules that all work together to produce a fully
functioning solution that is also fully testable, can be worked on by multiple teams
simultaneously, and is much easier to read, understand, and document.

We will finish the chapter off by looking at some programming design guidelines, mainly,
KISS, YAGNI, DRY, SOLID, and Occam's Razor.
Coding Standards and Principles in C# Chapter 1

The following topics will be covered in this chapter:

The need for coding standards, principles, and methodologies

Naming conventions and methods
Comments and formatting
Modularity
KISS
YAGNI
DRY
SOLID
Occam's Razor

The learning objectives for this chapter are for you to do the following:

Understand why bad code negatively impacts projects.

Understand how good code positively impacts projects.
Understand how coding standards improve code and how to enforce them.
Understand how coding principles enhance software quality.
Understand how methodologies aid the development of clean code.
Implement coding standards.
Choose solutions with the least assumptions.
Reduce code duplication and write SOLID code.

Technical requirements
To work on the code in this chapter, you will need to download and install Visual Studio
2019 Community Edition or higher. This IDE can be downloaded from https://
visualstudio.microsoft.com/.

You will find the code for this book located at https://github.com/PacktPublishing/
Clean-Code-in-C-. I have put them all under a single solution with each chapter as a
solution folder. You will find the code for each chapter in the relevant chapter folder. If
running a project, remember to assign it as the startup project.

[8]
Coding Standards and Principles in C# Chapter 1

Good code versus bad code

Both good code and bad code compile. That's the first thing to understand. The next thing
to understand is that bad code is bad for a reason, and likewise, good code is good for a
reason. Let's have a look at some of those reasons in the following comparison table:

Good Code Bad Code

Proper indentation. Improper indentation.
Meaningful comments. Comments that state the obvious.
Comments that excuse bad code.
API documentation comments.
Commented out lines of code.
Proper organization using namespaces. An improper organization using namespaces.
Good naming conventions. Bad naming conventions.
Classes that do one job. Classes that do multiple jobs.
Methods that do one thing. Methods that do many things.
Methods with less than 10 lines, and preferably no more
Methods with more than 10 lines of code.
than 4.
Methods with no more than two parameters. Methods with more than two parameters.
Proper use of exceptions. Using exceptions to control program flow.
Code that is readable. Code that is difficult to read.
Code that is loosely coupled. Code that is tightly coupled.
High cohesion. Low cohesion.
Objects are cleanly disposed of. Objects left hanging around.
Avoidance of the Finalize() method. Use of the Finalize() method.
The right level of abstraction. Over-engineering.
Use of regions in large classes. Lack of regions in large classes.
Encapsulation and information hiding. Directly exposing information.
Object-oriented code. Spaghetti code.
Design patterns. Design anti-patterns.

That's quite an exhaustive list, isn't it? In the following sections, we will look at how these
features and the differences between good and bad code impact the performance of your
code.

Bad code
We will now take a brief look at each of the bad coding practices that we listed earlier,
detailing specifically how that practice affects your code.

[9]
Coding Standards and Principles in C# Chapter 1

Improper indentation
Improper indentation can work toward making code really hard to read, especially if the
methods are large. For code to be easy to read by humans, we need proper indentation. If
code lacks proper indentation it can be very hard to see which part of the code belongs to
which block.

By default, Visual Studio 2019 correctly formats and indents your code when parentheses
and braces are closed. But sometimes, it incorrectly formats the code, to bring to your
attention that the code you've written contains an exception. But if you are using a simple
text editor, then you will have to do your formatting by hand.

Incorrectly indented code is also time-consuming to correct, and a frustrating waste of

programming time when it could easily have been avoided. Let's look at a simple code
example:
public void DoSomething()
{
for (var i = 0; i < 1000; i++)
{
var productCode = $"PRC000{i}";
//...implementation
}
}

The preceding code does not look all that nice, yet it is still readable. But the more lines of
code you add, the harder the code becomes to read.

It is very easy to miss a closing bracket. If your code is not properly indented, then this can
make finding the missing bracket that much harder, as you can not easily spot which code
block is missing its closing bracket.

Comments that state the obvious

I've seen programmers get really upset at comments that state the obvious as they find
them patronizing. In programming discussions that I have been part of, programmers have
stated how they dislike comments, and how they believe the code should be self-
documenting.

[ 10 ]
Coding Standards and Principles in C# Chapter 1

I can understand their sentiments. If you can read code without comments like you can
read a book and understand it, then it is a really good piece of code. If you have a variable
declared as a string, then why add a comment such as // string? Let's look at an
example:
public int _value; // This is used for storing integer values.

We know here that the value holds an integer by its type of int. So there really is no need
to state the obvious. All you're doing is wasting time and energy and cluttering up the
code.

Comments that excuse bad code

You may have a tight deadline to meet, but comments such as // I know this code
sucks but hey at least it works! are just awful. Don't do it. It shows a lack of
professionalism and can really disgruntle fellow programmers.

If you really are pushed to get something working out the door, raise a refactor ticket and
add it as part of a TODO comment such as // TODO: PBI23154 Refactor Code to
meet company coding practices. Then you or the other developers who are assigned
to work on technical debt can pick up the Product Backlog Item (PBI) and refactor the
code.

Here's another example:

...
int value = GetDataValue(); // This sometimes causes a divide by zero
error. Don't know why!
...

This one is really bad. Okay, thank you for letting us know that divide-by-zero errors occur
here. But have you raised a bug ticket? Have you tried to get to the bottom of it and fix it? If
everybody who is actively working on the project does not touch that code, how will they
know that buggy code is there?

At the very minimum, you should at least have a // TODO: comment in place. Then at
least the comment will show up in the Task List so that developers can be notified and
work on it.

[ 11 ]
Coding Standards and Principles in C# Chapter 1

Commented-out lines of code

If you comment out lines of code to try something, fine. But if you are going to use the
replacement code instead of the commented-out code, then delete the commented-out code
before you check it in. One or two commented outlines is not that bad. But when you have
many lines of commented-out code, it becomes distracting and makes code hard to
maintain; it can even lead to confusion:
/* No longer used as has been replaced by DoSomethinElse().
public void DoSomething()
{
// ...implementation...
}
*/

Why? Just why? If it has been replaced and is no longer needed, then just delete it. If your
code is in version control, and you need to get the method back, then you can always view
the history of the file and get the method back.

Improper organization of namespaces

When using namespaces, do not include code that should be elsewhere. This can make
finding the right code pretty hard or impossible, especially in large code bases. Let's look at
this example:
namespace MyProject.TextFileMonitor
{
+ public class Program { ... }
+ public class DateTime { ... }
+ public class FileMonitorService { ... }
+ public class Cryptography { ... }
}

We can see that all classes in the preceding code are under one namespace. Yet, we have the
opportunity to add three further namespaces to better organize this code:

MyProject.TextFileMonitor.Core: Core classes that define commonly used

members will be placed here, such as our DateTime class.
MyProject.TextFileMonitor.Services: All classes that act as a service will
be placed in this namespace, such as FileMonitorService.
MyProject.TextFileMonitor.Security: All security-related classes will be
placed in this namespace, including the Cryptography class in our example.

[ 12 ]
Coding Standards and Principles in C# Chapter 1

Bad naming conventions

In the days of Visual Basic 6 programming, we used to use Hungarian Notation. I
remember using it when I first switched to Visual Basic 1.0. It is no longer necessary to use
Hungarian Notation. Plus, it makes your code look ugly. So instead of using names such as
lblName, txtName, or btnSave, the modern way is to use NameLabel, NameTextBox,
and SaveButton, respectively

The use of cryptic names and names that don't seem to match the intention of the code can
make reading code rather difficult. What does ihridx mean? It means Human Resources
Index and is an integer. Really! Avoid using names such as mystring, myint, and
mymethod. Such names really don't serve a purpose.

Don't use underscores between words in a name either, such as Bad_Programmer. This can
cause visual stress for developers and can make the code hard to read. Simply remove the
underscore.

Don't use the same code convention for variables at the class level and method level. This
can make it difficult to establish the scope of a variable. A good convention for variable
names is to use camel case for variable names such as alienSpawn, and Pascal case for
method, class, struct, and interface names such as EnemySpawnGenerator.

Following the good variable name convention, you should distinguish between local
variables (those contained within a constructor or method), and member variables (those
placed at the top of the class outside of constructors and methods) by prefixing the member
variables with an underscore. I have used this as a coding convention in the workplace, and
it does work really well and programmers do seem to like this convention.

Classes that do multiple jobs

A good class should only do one job. Having a class that connects to a database, gets data,
manipulates that data, loads a report, assigns the data to the report, displays the report,
saves the report, prints the reports, and exports the report is doing too much. It needs to be
refactored into smaller, better-organized classes. All-encompassing classes like this are a
pain to read. I personally find them daunting. If you come across classes like this, organize
the functionality into regions. Then move the code in those regions into new classes that
perform one job.

Let's have a look at an example of a class that is doing multiple things:

public class DbAndFileManager
{
#region Database Operations

[ 13 ]
Coding Standards and Principles in C# Chapter 1

public void OpenDatabaseConnection() { throw new

NotImplementedException(); }
public void CloseDatabaseConnection() { throw new
NotImplementedException(); }
public int ExecuteSql(string sql) { throw new
NotImplementedException(); }
public SqlDataReader SelectSql(string sql) { throw new
NotImplementedException(); }
public int UpdateSql(string sql) { throw new
NotImplementedException(); }
public int DeleteSql(string sql) { throw new
NotImplementedException(); }
public int InsertSql(string sql) { throw new
NotImplementedException(); }

#endregion

#region File Operations

public string ReadText(string filename) { throw new

NotImplementedException(); }
public void WriteText(string filename, string text) { throw new
NotImplementedException(); }
public byte[] ReadFile(string filename) { throw new
NotImplementedException(); }
public void WriteFile(string filename, byte[] binaryData) { throw new
NotImplementedException(); }

#endregion
}

As you can see in the preceding code, the class does two main things: it performs database
operations and it performs file operations. Now the code is neatly organized within
correctly named regions used to logically separate code within a class. But the Single
Responsibility Principle (SRP) is broken. We would need to begin by refactoring this code
to separate out the database operations into a class of their own, called something like
DatabaseManager.

Then, we would remove the database operations from the DbAndFileManager class,
leaving only the file operations, and then rename the DbAndFileManager class to
FileManager. We would also need to consider the namespace of each file, and whether it
should be modified so that the DatabaseManager would be placed in the Data namespace
and the FileManager would be placed in the FileSystem namespace, or their equivalents
in your program.

[ 14 ]
Coding Standards and Principles in C# Chapter 1

The following code is the result of extracting the database code from the
DbAndFileManager class into its own class and in the correct namespace:

using System;
using System.Data.SqlClient;

namespace CH01_CodingStandardsAndPrinciples.GoodCode.Data
{
public class DatabaseManager
{
#region Database Operations

public void OpenDatabaseConnection() { throw new

#endregion
}
}

The refactoring of the filesystem code results in the FileManager class in the FileSystem
namespace, as shown in the following code:
using System;

namespace CH01_CodingStandardsAndPrinciples.GoodCode.FileSystem
{
public class FileManager
{
#region File Operations

public string ReadText(string filename) { throw new

[ 15 ]
Coding Standards and Principles in C# Chapter 1

public void WriteFile(string filename, byte[] binaryData) { throw

new NotImplementedException(); }

#endregion
}
}

We've seen how to identify classes that do too much, and how we can refactor them to do
only a single thing. Now let's repeat the process as we look at methods that do many things.

Methods that do many things

I have found myself getting lost in methods with many, many levels of indentation doing
many things in those various indentations. The permutations were mind-boggling. I
wanted to refactor the code to make maintenance easier, but my senior prohibited it. I could
clearly see how the method could have been smaller by farming out the code to different
methods.

Time for an example. In this example, the method accepts a string. That string is then
encrypted and decrypted. It is also long so that you can see why methods should be kept
small:
public string security(string plainText)
{
try
{
byte[] encrypted;
using (AesManaged aes = new AesManaged())
{
ICryptoTransform encryptor = aes.CreateEncryptor(Key, IV);
using (MemoryStream ms = new MemoryStream())
using (CryptoStream cs = new CryptoStream(ms, encryptor,
CryptoStreamMode.Write))
{
using (StreamWriter sw = new StreamWriter(cs))
sw.Write(plainText);
encrypted = ms.ToArray();
}
}
Console.WriteLine($"Encrypted data:
{System.Text.Encoding.UTF8.GetString(encrypted)}");
using (AesManaged aesm = new AesManaged())
{
ICryptoTransform decryptor = aesm.CreateDecryptor(Key, IV);
using (MemoryStream ms = new MemoryStream(encrypted))
{

[ 16 ]
Coding Standards and Principles in C# Chapter 1

using (CryptoStream cs = new CryptoStream(ms, decryptor,

CryptoStreamMode.Read))
{
using (StreamReader reader = new StreamReader(cs))
plainText = reader.ReadToEnd();
}
}
}
Console.WriteLine($"Decrypted data: {plainText}");
}
catch (Exception exp)
{
Console.WriteLine(exp.Message);
}
Console.ReadKey();
return plainText;
}

As you can see in the preceding method, it has 10 lines of code and is hard to read. Plus, it
is doing more than one thing. This code can be broken down into two methods that each
perform a single task. One method would encrypt a string, and the other method
would decrypt the string. This leads us nicely into why methods should have no more than
10 lines of code.

Methods with more than 10 lines of code

Large methods are not nice to read and understand. They can also lead to very hard-to-find
bugs. Another problem with large methods is they can lose sight of their original intent. It's
even worse when you come across large methods that have sections separated by
comments and code wrapped in regions.

If you have to scroll to read a method, then it is too long and can lead to programmer stress
and misinterpretation. This in turn can lead to modifications that will break the code or the
intent, or both. Methods should be as small as you can make them. But common sense does
need to be exercised, as you can take the matter of small methods to the nth degree to the
point that it becomes excessive. The key to getting the right balance is to ensure the intent
of the method is very clear and succinctly implemented.

The previous code is a good candidate for why you should keep methods small. Small
methods are easy to read and understand. Normally, if your code drifts beyond 10 lines it
may be doing more than it is intended to. Make sure your methods name their intentions,
as in OpenDatabaseConnection() and CloseDatabaseConnection(), and that they
stick to their intentions and do not deviate away from them.

[ 17 ]
Coding Standards and Principles in C# Chapter 1

We are now going to take a look at method parameters.

Methods with more than two parameters

Methods with many parameters tend to get a bit unwieldy. Apart from being hard to read,
it is very easy to pass a value to the wrong parameter and break type safety.

Testing methods get increasingly more complex as the number of parameters increases, the
main reason being that you have more permutations to apply to your test cases. It is
possible that you will miss a use case that will cause issues in production.

Using exceptions to control program flow

Exceptions used to control program flow may hide the intention of the code. They can also
lead to unexpected and unintended results. The very fact that your code has been
programmed to expect one or more exceptions shows your design to be wrong. A typical
scenario that is covered in more detail in Chapter 5, Exception Handling.

A typical scenario is when a business uses Business Rule Exceptions (BREs). A method
will perform an action anticipating that an exception will be thrown. The program flow will
be determined by whether the exception is thrown or not. A much better way is to use
available language constructs to perform validation checks that return a Boolean value.

The following code shows the use of a BRE to control program flow:
public void BreFlowControlExample(BusinessRuleException bre)
{
switch (bre.Message)
{
case "OutOfAcceptableRange":
DoOutOfAcceptableRangeWork();
break;
default:
DoInAcceptableRangeWork();
break;
}
}

The method accepts BusinessRuleException. Depending upon the message in the

exception, BreFlowControlExample() either calls the
DoOutOfAcceptableRangeWork() method or the DoInAcceptableRangeWork()
method.

[ 18 ]
Coding Standards and Principles in C# Chapter 1

A much better way to control the flow is through Boolean logic. Let's look at the
following BetterFlowControlExample() method:
public void BetterFlowControlExample(bool isInAcceptableRange)
{
if (isInAcceptableRange)
DoInAcceptableRangeWork();
else
DoOutOfAcceptableRangeWork();
}

In the BetterFlowControlExample() method, a Boolean value is passed into the

method. The Boolean value is used to determine which path to execute. If the condition is in
the acceptable range, then DoInAcceptableRangeWork() is called. Otherwise, the
DoOutOfAcceptableRangeWork() method is called.

Next, we will consider code that is difficult to read.

Code that is difficult to read

Code such as lasagna and spaghetti code is really hard to read or follow. Badly named
methods can also be a pain as they can obfuscate the intention of the method. Methods are
further obfuscated if they are large and if linked methods are separated by a number of
unrelated methods.

Lasagna code, also known more commonly as indirection, refers to layers of abstraction
where something is referred to by name rather than by action. Layering is used extensively
in Object-Oriented Programming (OOP) and to good effect. However, the more
indirection is used, the more complex code can become. This can make it very hard for new
programmers on a project to get up to speed with understanding the code. So there must be
a balance struck between indirection and ease of understanding.

Spaghetti code refers to a tangled mess of tightly coupled code with low cohesion. Such
code is very hard to maintain, refactor, extend, and redesign. Though on the plus side, it
can be very easy to read and follow since it is more procedural in its programming. I
remember working as a junior programmer on a VB6 GIS program that was sold to
companies and used for marketing purposes. My technical director and his senior
programmers had previously tried to redesign the software and failed. So they passed the
gauntlet to me so that I would redesign the program. But not being skilled in software
analysis and design at the time, I also failed.

[ 19 ]
Coding Standards and Principles in C# Chapter 1

The code was just too complex to follow and group into related items, and it was way too
big. With hindsight, I would have been better off making a list of everything the program
did, grouping the list by features, and then coming up with a list of requirements without
even looking at the code.

So my lesson learned when redesigning software is to avoid looking at the code at all costs.
Write down everything the program does, and what the new functionality is that it should
include. Turn the list into a set of software requirements with associated tasks, tests, and
acceptance criteria, and then program to the specifications.

Code that is tightly coupled

Code that is tightly coupled is hard to test and hard to extend or modify. It is also hard to
reuse code that is dependent on other code within a system.

An example of tight coupling is when you reference a concrete class type in the parameter
rather than referencing an interface. When referencing a concrete class, any changes to the
concrete class directly affect the class that references it. So if you have a database
connection class for a client that connects to SQL Server, and then takes on another
customer that requires an Oracle database, then the concrete class would have to be
modified for that specific customer and their Oracle database. That would lead to two
versions of the code.

The more customers there are, the more versions of the code required. This soon becomes
untenable and a right nightmare to maintain. Imagine that your database connection class
has 100,000 different clients using 1 of 30 variations of the class, and they all have the same
bug that has been identified and affects them all. That is 30 classes that have to have the
same fix put in place, tested, packaged, and deployed. That's a lot of maintenance
overhead, and very costly financially.

This particular scenario can be overcome by referencing an interface type and then using a
database factory to build the required connection object. Then the connection string can be
set in a configuration file by the customer and passed into the factory. The factory would
then produce a concrete connection class that implements a connection interface for the
specific type of database specified in the connection string.

Here is a bad example of tightly coupled code:

public class Database
{
private SqlServerConnection _databaseConnection;

public Database(SqlServerConnection databaseConnection)

[ 20 ]
Coding Standards and Principles in C# Chapter 1

{
_databaseConnection = databaseConnection;
}
}

As you can see from the example, our database class is tied to using SQL Server and would
require a hardcoded change to accept any other type of database. We will be covering
refactoring of code in later chapters with actual code examples.

Low cohesion
Low cohesion consists of unrelated code that performs a variety of different tasks all
grouped together. An example would be a utility class that contains a number of different
utility methods for handling dates, text, numbers, doing file input and output, data
validation, and encryption and decryption.

Objects left hanging around

When objects are left hanging around in memory, they can lead to memory leaks.

Static variables can lead to memory leaks in several ways. If you're not
using DependencyObject or INotifyPropertyChanged, then you are effectively
subscribing to events. The Common Language Runtime (CLR) creates a strong reference
by using the ValueChanged event via the PropertyDescriptors AddValueChanged
event, which results in the storage of PropertyDescriptor that references the object it is
bound to.

Unless you unsubscribe your bindings, you will end up with a memory leak. You will also
end up with memory leaks using static variables that reference objects that don't get
released. Any object that is referenced by a static variable is marked as not to be collected
by the garbage collector. This is because static variables that reference objects are Garbage
Collection (GC) roots, and anything that is a GC root is marked by the garbage collector as
do not collect.

When you use anonymous methods that capture class members, the instance of the class is
referenced. This causes a reference to the class instance to remain alive while the
anonymous methods stay alive.

When using unmanaged code (COM), if you do not release any managed and unmanaged
objects and explicitly deallocate any memory, then you will end up with memory leaks.

[ 21 ]
Coding Standards and Principles in C# Chapter 1

Code that caches indefinitely without using weak references, deleting unused cache, or
limiting the cache size will eventually run out of memory.

You would also end up with a memory leak if you were to create object references in a
thread that never terminates.

Event subscriptions that are not anonymous reference classes. While these events remain
subscribed to, the objects will remain in memory. So unless you unsubscribe from events
when they are not needed, it is likely you will end up with a memory leak.

Use of the Finalize() method

While finalizers can help free up resources from objects that have not been correctly
disposed of and help to prevent memory leaks, they do have a number of drawbacks.

You do not know when finalizers will be called. They will be promoted by the garbage
collector along with all dependants on the graph to the next generation, and will not be
garbage-collected until the garbage collector decides to do so. This can mean that objects
stay in memory for a long time. Out-of-memory exceptions could occur using finalizers as
you can be creating objects faster then they are getting garbage-collected.

Over-engineering
Over-engineering can be an utter nightmare. The biggest reason for this is that as a mere
human, wading through a massive system, trying to understand it, how you are to use it,
and what goes where is a time-consuming process. All the more so when there is no
documentation, you are new to the system, and even people who have been using it much
longer than you are unable to answer your questions.

This can be a major cause of stress when you are expected to work on it with set deadlines.

Learn to Keep It Simple, Stupid

A good example of this is at one of the places I've worked. I had to write a test for a web
app that accepted JSON from a service, allowed a child to do a test, and then passed the
resulting scoring to another service. I did not use OOP, SOLID, or DRY, as I should have
according to company policy. But I did get the work done by using KISS and procedural
programming with events in a very small time frame. I was penalized for it and forced to
rewrite it using their homegrown test player.

[ 22 ]
Coding Standards and Principles in C# Chapter 1

So I set about learning their test player. There was no documentation, it did not follow their
DRY principles, and very few people if any really understood it. Instead of a few days, like
my penalized system, my new version that had to use their system took weeks to build
because it did not do what I needed it to do, and I was not allowed to modify it to do what I
needed it to do. So I was slowed down while I waited for someone to do what was
required.

My first solution satisfied the business requirements and was an independent piece of code
that cared about nothing else. The second solution satisfied the development team's
technical requirements. The project lasted longer than the deadline. Any project that
overshoots its deadline costs the business more money than planned.

The other point I would like to make with my penalized system was that it was far simpler
and easier to understand than the newer system that was rewritten to use the generic test
player.

You don't always have to follow OOP, SOILD, and DRY. Sometimes it pays not to. After all,
you can write the most beautiful OOP system. But under the hood, your code is converted
to procedural code that is closer to what the computer understands!

Lack of regions in large classes

Large classes with lots of regions are very hard to read and follow, especially when related
methods are not grouped together. Regions are very good for grouping similar to members
within a large class. But they are no good if you don't use them!

Lost-intention code
If you are viewing a class and it is doing several things, then how do you know what its
original intention was? If you are looking for a date method, for example, and you find it in
a file class in the input/output namespace of your code, is the date method in the right
location? No. Will it be hard for other developers who don't know your code to find that
method? Of course it will. Take a look at this code:
public class MyClass
{
public void MyMethod()
{
// ...implementation...
}

public DateTime AddDates(DateTime date1, DateTime date2)

[ 23 ]
Coding Standards and Principles in C# Chapter 1

{
//...implementation...
}

public Product GetData(int id)

{
//...implementation...
}
}

What is the purpose of the class? The name does not give any indication, and what does
MyMethod do? The class also appears to be doing date manipulation and getting product
data. The AddDates method should be in a class solely for managing dates. And the
GetData method should be in the product's view model.

Directly exposing information

Classes that directly expose information are bad. Apart from producing tight coupling that
can lead to bugs, if you want to change the information type, you have to change the type
everywhere it is used. Also, what if you want to perform data validation before the
assignment? Here's an example:
public class Product
{
public int Id;
public int Name;
public int Description;
public string ProductCode;
public decimal Price;
public long UnitsInStock
}

In the preceding code, if you wanted to change UnitsInStock from type long to type int,
you would have to change the code everywhere it is referenced. You would have to do the
same with ProductCode. If new product codes had to adhere to a strict format, you would
not be able to validate product codes if the string could be directly assigned by the calling
class.

Good code
Now that you know what not to do, it's time to look briefly at some good coding practices
to be able to write pleasing, performant code.

[ 24 ]
Coding Standards and Principles in C# Chapter 1

Proper indentation
When you use proper indentation, it makes reading the code much easier. You can tell by
the indentation where code blocks start and end, and what code belongs to those code
blocks:
public void DoSomething()
{
for (var i = 0; i < 1000; i++)
{
var productCode = $"PRC000{i}";
//...implementation
}
}

In the preceding simple example, the code looks nice and is readable. You can clearly see
where each code block starts and finishes.

Meaningful comments
Meaningful comments are comments that express the programmer's intention. Such
comments are useful when the code is correct but may not be easily understood by anyone
new to the code, or even to the same programmer in a few week's time. Such comments can
be really helpful.

API documentation comments

A good API is an API that has good documentation that is easy to follow. API comments
are XML comments that can be used to generate HTML documentation. HTML
documentation is important for developers wanting to use your API. The better the
documentation, the more developers are likely to want to use your API. Here's an example:
/// <summary>
/// Create a new <see cref="KustoCode"/> instance from the text and
globals. Does not perform
/// semantic analysis.
/// </summary>
/// <param name="text">The code text</param>
/// <param name="globals">
/// The globals to use for parsing and semantic analysis. Defaults to
<see cref="GlobalState.Default"/>
/// </param>.
public static KustoCode Parse(string text, GlobalState globals = null) {
... }

[ 25 ]
Coding Standards and Principles in C# Chapter 1

This excerpt from the Kusto Query Language project is a good example of an API
documentation comment.

Proper organization using namespaces

Code that is properly organized and placed in appropriate namespaces can save developers
a good amount of time when looking for a particular piece of code. For instance, if you are
looking for classes and methods to do with dates and times, it would be a good idea to have
a namespace called DateTime, a class called Time for time-related methods, and a class
called Date for date-related methods.

The following is an example of the proper organization of namespaces:

Name Description

CompanyName.IO.FileSystem The namespace contains classes that define file and directory
operations.
The namespace contains classes for performing various conversion
CompanyName.Converters
operations.
The namespace contains types for managing stream input and
CompanyName.IO.Streams
output.

Good naming conventions

It is good to follow the Microsoft C# naming conventions. Use Pascal casing for
namespaces, classes, interfaces, enums, and methods. Use camel case for variable names
and argument names, and make sure to prefix member variables with an underscore.

Have a look at this example code:

using System;
using System.Text.RegularExpressions;

namespace CompanyName.ProductName.RegEx
{
/// <summary>
/// An extension class for providing regular expression extensions
/// methods.
/// </summary>
public static class RegularExpressions
{
private static string _preprocessed;

public static string RegularExpression { get; set; }

public static bool IsValidEmail(this string email)

[ 26 ]
Coding Standards and Principles in C# Chapter 1

{
// Email address: RFC 2822 Format.
// Matches a normal email address. Does not check the
// top-level domain.
// Requires the "case insensitive" option to be ON.
var exp = @"\A(?:[a-z0-9!#$%&'*+/=?^_`{|}~-]+(?:\.
[a-z0-9!#$%&'*+/=?^_`{|}~-]+)@(?:[a-z0-9](?:[a-z0-9-]
[a-z0-9])?\.)+[a-z0-9](?:[a-z0-9-]*[a-z0-9])?)\Z";
bool isEmail = Regex.IsMatch(email, exp, RegexOptions.IgnoreCase);
return isEmail;
}

// ... rest of the implementation ...

}
}

It shows suitable examples of naming conventions for namespaces, classes, member

variables, classes, parameters, and local variables.

Classes that only do one job

A good class is a class that does only one job. When you read the class, its intention is clear.
Only the code that should be in that class is in that class and nothing else.

Methods that do one thing

Methods should only do one thing. You should not have a method that does more than one
thing, such as decrypting a string and performing string replacement. A method's intent
should be clear. Methods that do only one thing are more inclined to be small, readable,
and intentional.

Methods with less than 10 lines, and preferably no more

than 4
Ideally, you should have methods that are no longer than 4 lines of code. However, this is
not always possible, so you should aim to have methods that are no more than 10 lines in
length so that they are easy to read and maintain.

[ 27 ]
Coding Standards and Principles in C# Chapter 1

Methods with no more than two parameters

It is best to have methods with no parameters, but having one or two is okay. If you start
having more than two parameters, you need to think about the responsibility of your class
and methods: are they taking on too much? If you do need more than two parameters, then
you are better placed to pass an object.

Any method with more than two parameters can become difficult to read and follow.
Having no more than two parameters makes the code readable, and a single parameter that
is an object is way more readable than a method with several parameters.

Proper use of exceptions

Never use exceptions to control program flow. Handle common conditions that might
trigger exceptions in such a way that an exception will not be raised or thrown. A good
class is designed in such a way that you can avoid exceptions.

Recover from exceptions and/or release resources by using try/catch/finally exceptions.

When catching exceptions, use specific exceptions that may be thrown in your code, so that
you have more detailed information to log or assist in handling the exception.

Sometimes, using the predefined .NET exception types is not always possible. In such cases,
it will be necessary to produce your own custom exceptions. Suffix your custom exception
classes with the word Exception, and make sure to include the following three
constructors:

Exception(): Uses default values

Exception(string): Accepts a string message
Exception(string, exception): Accepts a string message and an inner
exception

If you have to throw exceptions, don't return error codes but return exceptions with
meaningful information.

Code that is readable

The more readable the code is, the more developers will enjoy working with it. Such code is
easier to learn and work with. As developers come and go on a project, newbies will be able
to read, extend, and maintain the code with little effort. Readable code is also less inclined
to be buggy and unsafe.

[ 28 ]
Coding Standards and Principles in C# Chapter 1

Code that is loosely coupled

Loosely coupled code is easier to test and refactor. You can also swap and change loosely
coupled code more easily if you need to. Code reuse is another benefit of loosely coupled
code.

Let's use our bad example of a database being passed a SQL Server connection. We could
make that same class loosely coupled by referencing an interface instead of a concrete type.
Let's have a look at a good example of the refactored bad example from earlier:
public class Database
{
private IDatabaseConnection _databaseConnection;

public Database(IDatabaseConnection databaseConnection)

{
_databaseConnection = datbaseConnection;
}
}

As you can see in this rather basic example, as long as the passed-in class implements the
IDatabaseConnection interface, we can pass in any class for any kind of database
connection. So if we find a bug in the SQL Server connection class, only SQL Server clients
are affected. That means the clients with different databases will continue to work, and we
only have to fix the code for SQL Server customers in the one class. This reduces the
maintenance overhead and so reduces the overall cost of maintenance.

High cohesion
Common functionality that is correctly grouped together is known to be highly cohesive.
Such code is easy to find. For example, if you look at the Microsoft
System.Diagnostics namespace, you will find that it only contains code that pertains to
diagnostics. It would not make sense to include collections and filesystem code in the
Diagnostics namespace.

Objects are cleanly disposed of

When using disposable classes, you should always call the Dispose() method to cleanly
dispose of any resources that are in use. This helps to negate the possibility of memory
leaks.

[ 29 ]
Coding Standards and Principles in C# Chapter 1

There are times when you may need to set an object to null for it to go out of scope. An
example would be a static variable that holds a reference to an object that you no longer
require.

The using statement is also a good clean way to use disposable objects, as when the object
is no longer in scope it is automatically disposed of, so you don't need to explicitly call the
Dispose() method. Let's have a look at the code that follows:

using (var unitOfWork = new UnitOfWork())

{
// Perform unit of work here.
}
// At this point the unit of work object has been disposed of.

The code defines a disposable object in the using statement and does what it needs to
between the opening and closing curly braces. The object is automatically disposed of
before the braces are exited. And so there is no need to manually call the Dispose()
method, because it is called automatically.

Avoiding the Finalize() method

When using unmanaged resources, it is best to implement the IDisposable interface and
avoid using the Finalize() method. There is no guarantee of when finalizers will run.
They may not always run in the order you expect or when you expect them to run. Instead,
it is better and more reliable to dispose of unmanaged resources in the Dispose() method.

The right level of abstraction

You have the right level of abstraction when you expose to the higher level only that which
needs exposure, and you do not get lost in the implementation.

If you find that you are getting lost in the implementation details, then you have over-
abstracted. If you find that multiple people have to work in the same class at the same time,
then you have under-abstracted. In both cases, refactoring would be needed to get the
abstraction to the right level.

[ 30 ]
Coding Standards and Principles in C# Chapter 1

Using regions in large classes

Regions are very useful for grouping items within a large class as they can be collapsed. It
can be quite daunting reading through a large class and having to jump back and forth
between methods, so grouping methods that call each other in the class is a good way to
group them. The methods can then be collapsed and expanded as needed when working on
a piece of code.

As you can see from what we have looked at so far, good coding practices make for code
that is far more readable and easier to maintain. We will now take a look at the need for
coding standards and principles along with some software methodologies such as SOLID
and DRY.

The need for coding standards, principles,

and methodologies
Most software today is written by multiple teams of programmers. As you know, we all
have our own unique ways of coding, and we all have some form of programming
ideology. You can easily find programming debates regarding various software
development paradigms. But the consensus is that it does make our lives easier as
programmers if we do all adhere to a given set of coding standards, principles, and
methodologies.

Let's review what we mean by these in a little more detail.

Coding standards
Coding standards set out several dos and don'ts that must be adhered to. Such standards
can be enforced through tools such as FxCop and manually via peer code reviews. All
companies have their own coding standards that must be adhered to. But what you will
find in the real world is that when the business expects a deadline to be met, those coding
standards can go out of the window as the deadline can become more important than the
actual code quality. This is usually rectified by adding any required refactoring to the bug
list as technical debt to be addressed after the release.

[ 31 ]
Coding Standards and Principles in C# Chapter 1

Microsoft has its own coding standards, and the majority of the time these are the adopted
standards that are modified to suit each business' needs. Here are some examples of coding
standards found online:

https://www.c-sharpcorner.com/UploadFile/ankurmalik123/C-Sharp-coding-
standards/
https://www.dofactory.com/reference/csharp-coding-standards
https://blog.submain.com/coding-standards-c-developers-need/

When people across teams or within the same team adhere to coding standards, your code
base becomes unified. A unified code base is much easier to read, extend, and maintain. It is
also likely to be less error-prone. And if errors do exist, they are more likely to be found
more easily, since the code follows a standard set of guidelines that all developers adhere
to.

Coding principles
Coding principles are a set of guidelines for writing high-quality code, testing and
debugging that code, and performing maintenance on the code. Principles can be different
between programmers and programming teams.

Even if you are a lone programmer, you will do yourself an honorable service by defining
your own coding principles and sticking to them. If you work in a team, then it is very
beneficial to all to agree on a set of coding standards to make working on shared code
easier.

Throughout this book, you will see examples of coding principles such as SOLID, YAGNI,
KISS, and DRY, all of which will be explained in detail. But for now, SOLID stands for
Single Responsibility Principle, Open-Closed Principle, Liskov Substitution, Interface
Segregation Principle, and Dependency Inversion Principle. YAGNI stands for You Ain't
Gonna Need It. KISS stands for Keep It Simple, Stupid, and DRY stands for Don't Repeat
Yourself.

Coding methodologies
Coding methodologies break down the process of developing software into a number of
predefined phases. Each phase will have a number of steps associated with it. Different
developers and development teams will have their own coding methodologies that they
follow. The main aim of coding methodologies is to streamline the process from the initial
concept, through the coding phase, to the deployment and maintenance phases.

[ 32 ]
Coding Standards and Principles in C# Chapter 1

In this book, you will become accustomed to Test-Driven Development (TDD) and
Behavioral-Driven Development (BDD) using SpecFlow, and Aspect-Oriented
Programming (AOP) using PostSharp.

Coding conventions
It is best to implement the Microsoft C# coding conventions. You can review them at
https://docs.microsoft.com/en-us/dotnet/csharp/programming-guide/inside-a-
program/coding-conventions.

By adopting Microsoft's coding conventions, you are guaranteed to write code in a formally
accepted and agreed-upon format. These C# coding conventions help people to focus on
reading your code and spend less time focusing on the layout. Basically, Microsoft's coding
standards promote best practices.

Modularity
Breaking large programs up into smaller modules makes a lot of sense. Small modules are
easy to test, are more readily reused, and can be worked on independently from other
modules. Small modules are also easier to extend and maintain.

A modular program can be divided into different assemblies and different namespaces
within those assemblies. Modular programs are also much easier to work on in team
environments as different modules can be worked on by different teams.

In the same project, code is modularized by adding folders that reflect namespaces. A
namespace must only contain code that is related to its name. So, for instance, if you have a
namespace called FileSystem, then types related to files and directories should be placed
in that folder. Likewise, if you have a namespace called Data, then only types related to
data and data sources should be located in that namespace.

Another beautiful aspect of correct modularization is that if you keep modules small and
simple, they are easy to read. Most of a coder's life apart from coding is spent reading and
understanding code. So the smaller and more correctly modularized the code is, then the
more easier it is to read and understand the code. This leads to a greater understanding of
the code and improves developer take-up and use of the code.

[ 33 ]
Coding Standards and Principles in C# Chapter 1

KISS
You may be the super genius of the computer programming world. You may be able to
produce code that is so sexy that other programmers can only stare at it in awe and end up
drooling on their keyboard. But do those other programmers know what the code is by just
looking at it? If you found that code in 10 weeks' time when you head deep into a mountain
of different code with deadlines to meet, would you be able to explain with absolute clarity
what your code does and the rationale behind your choice of coding method? And have
you considered that you may have to work on that code further down the road?

Have you ever programmed some code, gone away, and then looked at it more than a few
days later and thought to yourself, I didn't write this rubbish, did I? What was I thinking!? I
know I've been guilty of it and so have some of my ex-colleagues.

When programming code, it is essential to keep the code simple and in a human-readable
format that even newbie junior programmers can understand. Often juniors are exposed to
code to read, understand, and then maintain. The more complex the code, the longer it
takes for juniors to get up to speed. Even seniors can struggle with complex systems to the
point that they leave to find work elsewhere that's less taxing on the brain and their well-
being.

For example, if you are working on a simple website, ask yourself a few questions. Does it
really need to use microservices? Is the brownfield project you are working on really
complicated? Is it possible to simplify it to make it easier to maintain? When developing a
new system, what are the minimum number of moving parts you need to write a robust,
maintainable, and scalable solution that performs well?

YAGNI
YAGNI is a discipline in the agile world of programming that stipulates that a programmer
should not add any code until it is absolutely needed. An honest programmer will write
failing tests based on a design, then write just enough production code for the tests to work,
and finally, refactor the code to remove any duplication. Using the YAGNI software
development methodology, you keep your classes, methods, and overall lines of code to an
absolute minimum.

The primary goal of YAGNI is to prevent the over-engineering of software systems by

computer programmers. Do not add complexity if it is not needed. You must remember to
only write the code that you need. Don't write code that you don't need, and don't write
code for the sake of experimentation and learning. Keep experimental and learning code in
sandboxed projects specifically for those purposes.

[ 34 ]
Coding Standards and Principles in C# Chapter 1

DRY
I said Don't Repeat Yourself! If you find that you are writing the same code in multiple areas,
then this is a definite candidate for refactoring. You should look at the code to see if it can
be genericized and placed in a helper class for use throughout the system or in a library for
use by other projects.

If you have the same piece of code in multiple locations, and you find the code has a fault
and needs to be modified, you must then modify the code in other areas. In situations like
this, it is very easy to overlook code that requires modification. The result is code that gets
released with the problem fixed in some areas, but still existing in others.

That is why it is a good idea to remove duplicate code as soon as you encounter it, as it may
cause more problems further down the road if you don't.

SOLID
SOLID is a set of five design principles that intend to make software easier to understand
and maintain. Software code should be easy to read and extend without having to modify
portions of the existing code. The five SOLID design principles are as follows:

Single Responsibility Principle: Classes and methods should only perform a

single responsibility. All the elements that form a single responsibility should be
grouped together and encapsulated.
Open/Closed Principle: Classes and methods should be open for extension and
closed for modification. When a change to the software is required, you should
be able to extend the software without modifying any of the code.
Liskov Substitution: Your function has a pointer to a base class. It must be able
to use any class derived from the base class without knowing it.
Interface Segregation Principle: When you have large interfaces, the clients that
use them may not need all the methods. So, using the Interface Segregation
Principle (ISP), you extract out methods to different interfaces. This means that
instead of having one big interface, you have many small interfaces. Classes can
then implement interfaces with only the methods they need.
Dependency Inversion Principle: When you have a high-level module, it should
not be dependent upon any low-level modules. You should be able to switch
between low-level modules freely without affecting the high-level module that
uses them. Both high-level and low-level modules should depend upon
abstractions.

[ 35 ]
Coding Standards and Principles in C# Chapter 1

An abstraction should not depend upon details, but details should depend upon
abstractions.

When you declare variables, you should always use static types such as an interface or
abstract class. Concrete classes that implement the interface or inherit from the abstract
class can then be assigned to the variable.

Occam's Razor
Occam's Razor states the following: Entities should not be multiplied without necessity. To
paraphrase, this essentially means that the simplest solution is most likely the correct one. So, in
software development, the breaking of the principle of Occam's Razor is accomplished by
making unnecessary assumptions and employing the least simple solution to a software
problem.

Software projects are usually founded upon a collection of facts and assumptions. Facts are
easy to deal with but assumptions are something else. When coming up with a software
project solution to a problem, you normally discuss the problem and potential solutions as
a team. When choosing a solution, you should always choose the project with the least
assumptions as this will be the most accurate choice to implement. If there are a few
fair assumptions, the more assumptions you are having to make, the more likely it is that
your design solution is flawed.

A project with less moving parts has less that can go wrong with it. So, by keeping projects
small with as few entities as possible by not making assumptions unless they are necessary,
and only dealing with facts, you adhere to the principle of Occam's Razor.

Summary
In this chapter, you have had an introduction to good code and bad code and, hopefully,
you now understand why good code matters. You have also been provided with the link to
the Microsoft C# coding conventions so that you can follow Microsoft best practices for
coding (if you are not already doing so).

You have also briefly been introduced to various software methodologies including DRY,
KISS, SOLID, YAGNI, and Occam's Razor.

[ 36 ]
Coding Standards and Principles in C# Chapter 1

Using modularity, you have seen the benefits of modularizing code using namespaces and
assemblies. Such benefits include independent teams being able to work on independent
modules, and code reusability and maintainability.

In the next chapter, we will be looking at peer code reviews. They can be unpleasant at
times, but peer code reviews help to keep programmers in check by making sure they are
adhering to the company coding procedure.

Questions
1. What are some of the outcomes of bad code?
2. What are some of the outcomes of good code?
3. What are some of the benefits of writing modular code?
4. What is DRY code?
5. Why should you KISS when writing code?
6. What does the acronym SOLID stand for?
7. Explain YAGNI.
8. What's Occam's Razor?

Further reading
Adaptive Code: Agile coding with design patterns and SOLID principles, Second Edition
by Gary McLean Hall.
Hands-On Design Patterns with C# and .NET Core by Jeffrey Chilberto and Gaurav
Aroraa.
Building Maintainable Software, C# Edition by Rob can der Leek, Pascal can Eck,
Gijs Wijnholds, Sylvan Rigal, and Joost Visser.
Good information on software anti-patterns, including a long list of anti-patterns,
can be found at https://en.wikibooks.org/wiki/Introduction_to_Software_
Engineering/Architecture/Anti-Patterns.
Good information on design patterns, with a list of design patterns that links to
diagrams and implementation source code, can be found at https://en.
wikipedia.org/wiki/Software_design_pattern.

[ 37 ]
2
Code Review – Process and
Importance
The primary motivation behind any code review is to improve the overall quality of the
code. Code quality is very important. This almost goes without saying, especially if your
code is part of a team project or is accessible to others, such as open source developers and
customers through escrow agreements.

If every developer was free to code as they pleased, you would end up with the same kind
of code written in so many different ways, and ultimately the code would become an
unwieldy mess. That is why it is important to have a coding standards policy that outlines
the company's coding practices and code review procedures that are to be followed.

When code reviews are carried out, colleagues will review the code of other colleagues.
Colleagues will understand that it is only human to make mistakes. They will check the
code for mistakes, coding that breaks the company's code of coding conduct, and any code
that, while syntactically correct, can be improved upon to either make it more readable,
more maintainable, or more performant.
Code Review – Process and Importance Chapter 2

Therefore, in this chapter, we will cover the following topics to understand the code review
process in detail:

Preparing code for review

Leading a code review
Knowing what to review
Knowing when to send code for review
Providing and responding to review feedback

Please note that for the Preparing code for review and Knowing when to send
code for review sections, we will be talking from the point of view of
the programmer. For the Leading a code review and Knowing what to
review sections, we will be talking from the point of view of the code
reviewer. However, as regards the Providing and responding to review
feedback section, we will cover the viewpoints of both the programmer and
the code reviewer.

The learning objectives for this chapter are for you to be able to do the following:

Understand code reviews and why they are good

Partake in code reviews
Provide constructive criticism
Respond positively to constructive criticism

Before we dive deep into these topics, let's understand the general code review process.

The code review process

The normal procedure for carrying out a code review is to make sure your code compiles
and meets the requirements set. It should also pass all unit tests and end-to-end tests. Once
you are confident that you are able to compile, test, and run your code successfully, then it
is checked in to the current working branch. Once checked in, you will then issue a pull
request.

A peer reviewer will then review your code and share comments and feedback. If your
code passes the code review, your code review is completed and you can then merge your
working branch into the main trunk. Otherwise, the peer review will be rejected, and you
will be required to review your work and address the issues raised in the comments
provided by your reviewer.

[ 39 ]
Code Review – Process and Importance Chapter 2

The following diagram shows the peer code review process:

Preparing code for review

Preparing for a code review can be a right royal pain at times, but it does work for better
overall quality of code that is easy to read and maintain. It is definitely a worthwhile
practice that teams of developers should carry out as standard coding procedures. This is
an important step in the code review process, as perfecting this step can save the reviewer
considerable time and energy in performing the review.

Here are some standard points to keep in mind when preparing your code for review:

Always keep the code review in mind: When beginning any programming, you
should have the code review in mind. So keep your code small. If possible, limit
your code to one feature.
Make sure that all your tests pass even if your code builds: If your code builds
but you have failing tests, then deal immediately with what's causing those tests
to fail. Then, when the tests pass as expected, you can move on. It is important to
make sure that all unit tests are passed, and that end-to-end testing passes all
tests. It is important that all testing is complete and gets the green light, since
releasing code that works but was a test fail could result in some very unhappy
customers when the code goes to production.

[ 40 ]
Code Review – Process and Importance Chapter 2

Remember YAGNI: As you code, make sure to only add code that is necessary to
meet the requirement or feature you are working on. If you don't need it yet, then
don't code it. Only add code when it is needed and not before.
Check for duplicate code: If your code must be object-oriented and be DRY and
SOLID, then review your own code to see whether it contains any procedural or
duplicate code. Should it do so, take the time to refactor it so that it is object-
oriented, DRY, and SOLID.
Use static analyzers: Static code analyzers that have been configured to enforce
your company's best practices will check your code and highlight any issues that
are encountered. Make sure that you do not ignore information and warnings.
These could cause you issues further down the line.

Most importantly, only check your code in when you are confident that
your code satisfies business requirements, adheres to coding standards,
and passes all tests. If you check your code in as part of a Continuous
Integration (CI) pipeline, and your code fails the build, then you will
need to address the areas of concern raised by the CI pipeline. When you
are able to check in your code and the CI gives the green light, then you
can issue a pull request.

Leading a code review

When leading code reviews, it is important to have the right people present. The people
who will be in attendance at the peer code review will be agreed upon with the project
manager. The programmer(s) responsible for submitting the code for review will be present
at the code review unless they work remotely. In the case of remote working, the reviewer
will review the code and either accept the pull request, decline the pull request, or send the
developer some questions to be answered before taking any further action.

[ 41 ]
Code Review – Process and Importance Chapter 2

A suitable lead for a code review should possess the following skills and knowledge:

Be a technical authority: The person leading the code review should be a

technical authority that understands the company's coding guidelines and
software development methodologies. It is also important that they have a good
overall understanding of the software under review.
Have good soft skills: As the leader of the code review, the person must be a
warm and encouraging individual who is able to provide constructive feedback.
It is important that the person reviewing the programmer's code has good soft
skills so that there is no conflict between the reviewer and the person whose code
is being reviewed.
Not be overly critical: The leader of the code review must not be over-critical
and must be able to explain their critique of the programmer's code. It is useful if
the leader has been exposed to different programming styles, and can view the
code objectively to ensure that it meets the project's requirements.

In my experience, peer code reviews are always carried out on pull requests in the version
control tool being used by the team. A programmer will submit the code to version control
and then issue a pull request. The peer code reviewer will then review the code in the pull
request. Constructive feedback will be provided in the form of comments that will be
attached to the pull request. If there are problems with the pull request, then the reviewer
will reject the change request and comment on specific issues that need to be addressed by
the programmer. If the code review is successful, then the reviewer may add a comment
providing positive feedback, merge the pull request, and close it.

Programmers will need to note any comments made by the reviewer and take them on
board. If the code needs to be resubmitted, then the programmer will need to ensure that all
the reviewer's comments have been addressed prior to resubmitting.

It is a good idea to keep code reviews short, and to not review too many lines at any one
time.

Since a code review normally starts with a pull request, we will look at issuing a pull
request followed by responding to a pull request.

[ 42 ]
Code Review – Process and Importance Chapter 2

Issuing a pull request

When you have finished coding and you are confident in the quality of your code and that
it builds, you are able to then push or check in your changes, depending on what source
control system you use. When your code has been pushed, you can then issue a pull
request. When a pull request is raised, other people that are interested in the code are
notified and able to review your changes. These changes can then be discussed and
comments made regarding any potential changes that you need to make. In essence, your
pushing to your source control repository and issuing a pull request is what kick-starts the
peer code review process.

To issue a pull request, all you have to do (once you've checked your code in or pushed it)
is click on the Pull requests tab of your version control. There will then be a button you can
click on – New pull request. This will add your pull request to a queue to be picked up by
the relevant reviewers.

In the following screenshots, we will see the process of requesting and fulfilling a pull
request via GitHub:

1. On your GitHub project page, click on the Pull requests tab:

[ 43 ]
Code Review – Process and Importance Chapter 2

2. Then, click on the New pull request button. This will display the Comparing
changes page:

3. If you are happy, then click on the Create pull request button to start the pull
request. You will then be presented with the Open a pull request screen:

[ 44 ]
Code Review – Process and Importance Chapter 2

4. Write your comment regarding the pull request. Provide all the necessary
information for the code reviewer, but keep it brief and to the point. Useful
comments include identification of what changes have been made. Modify the
Reviewers, Assignees, Labels, Projects, and Milestones fields as necessary.
Then, once you are happy with the pull request details, click on the Create pull
request button to create the pull request. Your code will now be ready to be
reviewed by your peers.

[ 45 ]
Code Review – Process and Importance Chapter 2

Responding to a pull request

Since the reviewer is responsible for reviewing pull requests prior to merges of branches,
we would do well to look at responding to pull requests:

1. Start by cloning a copy of the code under review.

2. Review the comments and changes in the pull request.
3. Check that there are no conflicts with the base branch. If there are, then you will
have to reject the pull request with the necessary comments. Otherwise, you can
review the changes, make sure the code builds without errors, and make sure
there are no compilation warnings. At this stage, you will also look out for code
smells and any potential bugs. You will also check that the tests build, run, are
correct, and provide good test coverage of the feature to be merged. Make any
comments necessary and reject the pull request unless you are satisfied. When
satisfied, you can add your comments and merge the pull request by clicking on
the Merge pull request button, as shown here:

[ 46 ]
Code Review – Process and Importance Chapter 2

4. Now, confirm the merge by entering a comment and clicking on the Confirm
merge button:

5. Once the pull request has been merged and the pull request closed, the branch
can be deleted by clicking on the Delete branch button, as can be seen in the
following screenshot:

[ 47 ]
Code Review – Process and Importance Chapter 2

In the previous section, you saw how the reviewee raises a pull request to have their code
peer-reviewed before it is merged. And in this section, you have seen how to review a pull
request and complete it as part of a code review. Now, we will look at what to review in a
peer code review when responding to a pull request.

Effects of feedback on reviewees

When performing a code review of your peer's code, you must also take into consideration
the fact that feedback can be positive or negative. Negative feedback does not provide
specific details about the problem. The reviewer focuses on the reviewee and not on the
problem. Suggestions for improving the code are not offered to the reviewee by the
reviewer, and the reviewer's feedback is aimed at hurting the reviewee.

[ 48 ]
Code Review – Process and Importance Chapter 2

Such negative feedback received by the reviewee offends them. This has a negative impact
and can cause them to start doubting themselves. A lack of motivation then develops
within the reviewee and this can negatively impact the team, as work is not done on time or
to the required level. The bad feelings between the reviewer and the reviewee will also be
felt by the team, and an oppressive atmosphere that negatively impacts everyone on the
team can ensue. This can lead to other colleagues becoming demotivated, and the overall
project can end up suffering as a result.

In the end, it gets to the point where the reviewee has had enough and leaves for a new
position somewhere else to get away from it all. The project then suffers time-wise and
even financially, as time and money will need to be spent on finding a replacement.
Whoever is found to fill the position then has to be trained upon the system and the
working procedures and guidelines. The following diagram shows negative feedback from
the reviewer toward the reviewee:

Conversely, positive feedback from the reviewer to the reviewee has the opposite effect.
When the reviewer provides positive feedback to the reviewee, they focus on the problem
and not on the person. They explain why the code submitted is not good, along with the
problems it can cause. The reviewer will then suggest to the reviewee ways in which the
code can be improved. The feedback provided by the reviewer is only done to improve the
quality of the code submitted by the reviewee.

[ 49 ]
Code Review – Process and Importance Chapter 2

When the reviewee receives the positive (constructive) feedback, they respond in a positive
manner. They take on board the reviewer's comments and respond in the appropriate
manner by answering any questions, asking any relevant questions themselves, and the
code is then updated, based on the reviewer's feedback. The amended code is then
resubmitted for review and acceptance. This has a positive impact on the team as the
atmosphere remains a positive one, and work is done on time and to the required quality.
The following diagram shows the results of positive feedback on the reviewee from the
reviewer:

The point to remember is that your feedback can be constructive or destructive. Your aim as
a reviewer is to be constructive and not destructive. A happy team is a productive team. A
demoralized team is not productive and is damaging to the project. So, always strive to
maintain a happy team through positive feedback.

A technique for positive criticism is the feedback sandwich technique. You start with praise
on the good points, then you provide constructive criticism, and then you finish with
further praise. This technique can be very useful if you have members on the team that
doesn't react well to any form of criticism. Your soft skills in dealing with people are just as
important as your software skills in delivering quality code. Don't forget that!

We will now move on to look at what we should review.

[ 50 ]
Code Review – Process and Importance Chapter 2

Knowing what to review

There are different aspects of the code that have to be considered when reviewing it.
Primarily, the code being reviewed should only be the code that was modified by the
programmer and submitted for review. That's why you should aim to make small
submissions often. Small amounts of code are much easier to review and comment on.

Let's go through different aspects a code reviewer should assess for a complete and
thorough review.

Company's coding guidelines and business

requirement(s)
All code being reviewed should be checked against the company's coding guidelines and
the business requirement(s) the code is addressing. All new code should adhere to the latest
coding standards and best practices employed by the company.

There are different types of business requirements. These requirements include those of the
business and the user/stakeholder as well as functional and implementation requirements.
Regardless of the type of requirement the code is addressing, it must be fully checked for
correctness in meeting requirements.

For example, if the user/stakeholder requirement states that as a user, I want to add a new
customer account, does the code under review meet all the conditions set out in this
requirement? If the company's coding guidelines stipulate that all code must include unit
tests that test the normal flow and exceptional cases, then have all the required tests been
implemented? If the answer to any of these questions is no, then the code must be
commented on, the comments addressed by the programmer, and the code resubmitted.

Naming conventions
The code should be checked to see whether the naming conventions have been followed for
the various code constructs, such as classes, interfaces, member variables, local variables,
enumerations, and methods. Nobody likes cryptic names that are hard to decipher,
especially if the code base is large.

[ 51 ]
Code Review – Process and Importance Chapter 2

Here are a couple of questions that a reviewer should ask:

Are the names long enough to be human-readable and understandable?

Are they meaningful in relation to the intent of the code, but short enough to not
irritate other programmers?

As the reviewer, you must be able to read the code and understand it. If the code is difficult
to read and understand, then it really needs to be refactored before being merged.

Formatting
Formatting goes a long way to making code easy to understand. Namespaces, braces, and
indentation should be employed according to the guidelines, and the start and end of code
blocks should be easily identifiable.

Again, here is a set of questions a reviewer should consider asking in their review:

Is code to be indented using spaces or tabs?

Has the correct amount of white space been employed?
Are there any lines of code that are too long that should be spread over multiple
lines?
What about line breaks?
Following the style guidelines, is there only one statement per line? Is there only
one declaration per line?
Are continuation lines correctly indented using one tab stop?
Are methods separated by one line?
Are multiple clauses that make up a single expression separated by parentheses?
Are classes and methods clean and small, and do they only do the work they are
meant to do?

Testing
Tests must be understandable and cover a good subset of use cases. They must cover the
normal paths of execution and exceptional use cases. When it comes to testing the code, the
reviewer should check for the following:

Has the programmer provided tests for all the code?

Is there any code that is untested?
Do all the tests work?

[ 52 ]
Code Review – Process and Importance Chapter 2

Do any of the tests fail?

Is there adequate documentation of the code, including comments,
documentation comments, tests, and product documentation?
Do you see anything that stands out that, even if it compiles and works in
isolation, could cause bugs when integrated into the system?
Is the code well documented to aid maintenance and support?

Let's see how the process goes:

Untested code has the potential to raise unexpected exceptions during testing and
production. But just as bad as code that is not tested are tests that are not correct. This can
lead to bugs that are hard to diagnose, can be annoying for the customer, and make more
work for you further down the line. Bugs are technical debt and looked upon negatively by
the business. Moreover, you may have written the code, but others may have to read it as
they maintain and extend the project. It is always a good idea to provide some
documentation for your colleagues.

Now, concerning the customer, how are they going to know where your features are and
how to use them? Good documentation that is user-friendly is a good idea. And remember,
not all your users may be technically savvy. So, cater to the less technical person that may
need handholding, but do it without being patronizing.

As a technical authority reviewing the code, do you detect any code smells that may
become a problem? If so, then you must flag, comment, and reject the pull request and get
the programmer to resubmit their work.

[ 53 ]
Code Review – Process and Importance Chapter 2

As a reviewer, you should check that those exceptions are not used to control the program
flow and that any errors raised have meaningful messages that are helpful to developers
and to the customers who will receive them.

Architectural guidelines and design patterns

The new code must be checked to see whether it conforms to the architectural guidelines
for the project. The code should follow any coding paradigms that the company employs,
such as SOLID, DRY, YAGNI, and OOP. In addition, where possible, the code should
employ suitable design patterns.

This is where the Gang-of-Four (GoF) patterns come into play. The GOF comprises four
authors of a C++ book called Design Patterns: Elements of Reusable Object-Oriented Software.
The authors were Erich Gamma, Richard Helm, Ralph Johnson, and John Vlissides.

Today, their design patterns are heavily used in most, if not all, object-oriented
programming languages. Packt has books that cover design patterns, including .NET Design
Patterns, by Praseen Pai and Shine Xavier. Here is a really good resource that I recommend
that you visit: https://www.dofactory.com/net/design-patterns. The site covers each of
the GoF patterns and provides the definition, UML class diagram, participants, structural
code, and some real-world code for the patterns.

GoF patterns consist of creational, structural, and behavioral design patterns. Creational
design patterns include Abstract Factory, Builder, Factory Method, Prototype, and
Singleton. Structural design patterns include Adapter, Bridge, Composite, Decorator,
Façade, Flyweight, and Proxy. Behavioral design patterns include Chain of Responsibility,
Command, Interpreter, Iterator, Mediator, Memento, Observer, State, Strategy, Template
Method, and Visitor.

The code should also be properly organized and placed in the correct namespace and
module. Check the code also to see whether it is too simplistic or over-engineered.

Performance and security

Other things that may need to be considered include performance and security:

How well does the code perform?

Are there any bottlenecks that need to be addressed?

[ 54 ]
Code Review – Process and Importance Chapter 2

Is the code programmed in such a way to protect against SQL injection attacks
and denial-of-service attacks?
Is code properly validated to keep the data clean so that only valid data gets
stored in the database?
Have you checked the user interface, documentation, and error messages for
spelling mistakes?
Have you encountered any magic numbers or hard coded values?
Is the configuration data correct?
Have any secrets accidentally been checked in?

A comprehensive code review will encompass all of the preceding aspects and their
respective review parameters. But let's find out when it is actually the right time to even be
performing a code review.

Knowing when to send code for review

Code reviews should take place when the development is complete and before the
programmer of the code passes the code on to the QA department. Before any code is
checked into version control, all the code should build and run without errors, warnings, or
information. You can ensure this by doing the following:

You should run static code analysis on your programs to see whether any issues
are raised. If you receive any errors, warnings, or information, then address each
point raised. Do not ignore them as they can cause problems further down the
line. You can access the Code Analysis configuration dialog on the Code
Analysis page of the Visual Studio 2019 Project Properties tab. Right-click on
your project and select Properties | Code Analysis.
You should also make sure that all your tests run successfully, and you should
aim to have all your new code to be fully covered by normal and exceptional use
cases that test the correctness of your code against the specification you are
working on.
If you employ a continuous development software practice within your place of
work that integrates your code into a larger system, then you need to make sure
that the system integration is successful and that all tests run without failing. If
any errors are encountered, then you must fix them before you go any further.

[ 55 ]
Code Review – Process and Importance Chapter 2

When your code is complete, fully documented, and your tests work, and your system
integration all works without any issues, then that is the best time to undergo a peer code
review. Once you have reached the point that your peer code review is approved, your
code can then be passed on to the QA department. The following diagram shows the
Software Development Life Cycle (SDLC) from the development of the code through to
the end of the life of the code:

The programmer codes the software as per specifications. They submit the source code to
the version control repository and issue a pull request. The request is reviewed. If the
request fails, then the request is rejected with comments. If the code review passes, then the
code is deployed to the QA team that carry out their own internal testing. Any bugs found
are raised for the developers to fix. If the internal testing passes QA, then it is deployed into
User Acceptance Testing (UAT).

If UAT fails, then bugs are raised with the DevOps team, who could be developers or
infrastructure. If UAT passes QA, then it is deployed to staging. Staging is the team
responsible for deploying the product in the production environment. When the software is
in the hands of the customer, they raise a bug report if they encounter any bugs. Developers
then work on fixing the customer's bugs, and the process is restarted. Once the product
reaches the end of its life, it is retired from service.

[ 56 ]
Code Review – Process and Importance Chapter 2

Providing and responding to review

feedback
It is worth remembering that code reviews are aimed at the overall quality of code in
keeping with the company's guidelines. Feedback, therefore, should be constructive and
not used as an excuse to put down or embarrass a fellow colleague. Similarly, reviewer
feedback should not be taken personally and responses to the reviewer should focus on
suitable action and explanation.

The following diagram shows the process of issuing a Pull Request (PR), performing a
code review, and either accepting or rejecting the PR:

Providing feedback as a reviewer

Workplace bullying can be a problem, and programming environments are not immune.
Nobody likes a cocky programmer who thinks they are big. So, it is important that the
reviewer has good soft skills and is very diplomatic. Bear in mind that some people can
easily be offended and take things the wrong way. So know who you are dealing with and
how they are likely to respond; this will help you choose your approach and your words
carefully.

As the peer code reviewer, you will be responsible for understanding the requirements and
making sure the code meets that requirement. So, look for the answers to these questions:

Are you able to read and understand the code?

Can you see any potential bugs?

[ 57 ]
Code Review – Process and Importance Chapter 2

Have any trade-offs been made?

If so, why were the trade-offs made?
Do the trade-offs incur any technical debt that will need to be factored into the
project further down the line?

Once your review is complete, you will have three categories of feedback to choose from:
positive, optional, and critical. With positive feedback, you can provide commendations on
what the programmer has done really well. This is a good way to bolster morale as it can
often run low in programming teams. Optional feedback can be very useful in helping
computer programmers to hone their programming skills in line with the company
guidelines, and they can work to improve the overall wellbeing of the software being
developed.

Finally, we have critical feedback. Critical feedback is necessary for any problems that
have been identified and must be addressed before the code can be accepted and passed on
to the QA department. This is the feedback where you will need to choose your words
carefully to avoid offending anyone. It is important that your critical comments address the
specific issue being raised with valid reasons to support the feedback.

Responding to feedback as a reviewee

As the reviewee programmer, you must effectively communicate the background of your
code to your reviewer. You can help them by making small commits. Small amounts of
code are much easier to review than large amounts of code. The more code being reviewed,
the easier it is for things to be missed and slip through the net. While you are waiting for
your code to be reviewed, you must not make any further changes to it.

As you can guess, you will receive either positive, optional, or critical feedback from the
reviewer. The positive feedback works to boost your confidence in the project as well as
your morale. Build upon it and continue with your good practices. You may choose to act
or not upon optional feedback, but it's always a good idea to talk it through with your
reviewer.

For critical feedback, you must take it seriously and act upon it as this feedback is
imperative for the very success of the project. It is very important that you handle critical
feedback in a polite and professional manner. Don't allow yourself to be offended by any
comments from your reviewer; they are not meant to be personal. This is especially
important for new programmers, and programmers who lack confidence.

As soon as you receive your reviewer's feedback, act upon it, and make sure that you
discuss it with them as necessary.

[ 58 ]
Code Review – Process and Importance Chapter 2

Summary
In this chapter, we have discussed the importance of performing code reviews and the
complete process of getting code ready for review and responding to reviewer comments as
the programmer, along with how to lead a code review and what to look for when
performing a review as the code reviewer. It can be seen that there are clearly two roles in a
peer code review. These are the reviewer and the reviewee. The reviewer is the person
performing the code review, and the reviewee is the person whose code is being reviewed.

You have also seen how you, as a reviewer, can categorize your feedback and why soft
skills are important when providing feedback to fellow programmers. And as a reviewee
whose code is being scrutinized, you have seen how important it is to build upon positive
and optional feedback and how important it is to act upon critical feedback.

By now, you should have a good understanding of why it is important to conduct regular
code reviews, and why they should be done before the code is passed on to the QA
department. Peer code reviews do take time and can be uncomfortable for both the
reviewer and reviewee. But in the long run, they work toward a high-quality product that
is easy to extend and maintain, and they lead to better code reuse as well.

In the next chapter, we will be looking at how to write clean classes, objects, and data
structures. You will see how we can organize our classes, ensure our classes only have one
responsibility, and comment on our classes in order to assist with documentation
generation. We will then look at cohesion and coupling, designing for change, and the Law
of Demeter. Then, we will look at immutable objects and data structures, hiding data, and
exposing methods in objects, before finally looking at data structures.

Questions
1. What are the two roles involved in a peer code review?
2. Who agrees on the people that will be involved in the peer code review?
3. How can you save your reviewer time and effort prior to requesting a peer code
review?
4. When reviewing code, what kinds of things must you look out for?
5. What are the three categories of feedback?

[ 59 ]
Code Review – Process and Importance Chapter 2

Further reading
https://docs.microsoft.com/en-us/visualstudio/code-quality/?view=vs-
2019: This documentation by Microsoft provides information on the different
tools available to help you analyze and improve the quality and maintainability
of your code.
https://en.wikipedia.org/wiki/Code_review: There are many useful links on
this page to further your knowledge of code reviews and their value to your
business.
https://springframework.guru/gang-of-four-design-patterns/: Gang-of-
Four design patterns book.
https://www.packtpub.com/application-development/net-design-patterns:
.NET Design Patterns, by Praseed Pai and Shine Xavier.
https://help.github.com/en: GitHub's help page.

[ 60 ]
3
Classes, Objects, and Data
Structures
In this chapter, we will look at organizing, formatting, and commenting on classes. We will
also look at writing clean C# objects and data structures that respect the Law of Demeter. In
addition, we will look at immutable objects and data structures and the interfaces and
classes that define immutable collections in the System.Collections.Immutable
namespace.

We will cover the following broad topics:

Organizing classes
Commenting for document generation
Cohesion and coupling
The Law of Demeter
Immutable objects and data structures
Classes, Objects, and Data Structures Chapter 3

As you progress through this chapter, you will learn the following skills:

How to effectively organize your classes using namespaces.

Your classes will become smaller and more meaningful as you learn to program
them with only a single responsibility.
When it comes to writing your own APIs, you will be able to provide good
developer documentation by providing comments that aid document generation
tools.
Any programs you write will be easy to modify and extend due to their high
cohesion and low coupling.
Finally, you will be able to apply the Law of Demeter and write and use
immutable data structures.

So, let's start by looking at how we can effectively organize our classes by using
namespaces.

Technical requirements
You can access the code for this chapter on GitHub, at https://github.com/
PacktPublishing/Clean-Code-in-C-/tree/master/CH03.

Organizing classes
You will notice that the hallmark of a clean project is that it will have well-organized
classes. And folders will be used to group classes that belong together. Further, the classes
in the folders will be enclosed within namespaces that match the assembly name and folder
structure.

Each interface, class, struct, and enum should have its own source file in the correct
namespace. Source files should be logically grouped together in the appropriate folders and
the namespaces for the source files should match the assembly name and folder structure.
The following screenshot demonstrates a clean folder and file structure:

[ 62 ]
Classes, Objects, and Data Structures Chapter 3

[ 63 ]
Classes, Objects, and Data Structures Chapter 3

It is a bad idea to have more than one interface, class, struct, or enum in an
actual source file. The reason for this is that it can make locating items
difficult, despite the fact that we have IntelliSense to assist us.

When thinking about your namespaces, it is a good idea to follow the Pascal casing
sequence of company name, product name, technology name, and then plural names for
components separated by spaces. See the following for an example:
FakeCompany.Product.Wpf.Feature.Subnamespace {} // Product, technology and
feature specific.

The reason for starting with the company name is that it helps to avoid namespace classes.
So, if Microsoft and FakeCompany both have a namespace called System, which System
you desire to use can be differentiated by the company name.

Next, any items of code that are able to be reused in multiple projects are best placed in
separate assemblies that can be accessed by multiple projects:
FakeCompany.Wpf.Feature.Subnamespace {} /* Technology and feature specific.
Can be used across multiple products. */

When using tests in your code, such as when doing Test-Driven Development (TDD), it is
always best to keep your test classes in separate assemblies. Test assemblies should always
be given the name of the assembly they are testing with the namespace Tests appended to
the end of the assembly name:
FakeCompany.Core.Feature {} /* Technology agnostic and feature specific.
Can be used across multiple products. */

You should never put tests for different assemblies in the same test assembly as each other.
Always keep them separate.

In addition, the namespace and type should not use the same name as this can produce
compiler conflicts. When pluralizing namespaces, you can forego pluralizing for company
names, product names, and acronyms.

To summarize, here are the rules to keep in mind when organizing classes:

Follow the Pascal casing sequence of company name, product name, technology
name, and plural names for components separated by spaces.
Place reusable items of code in separate assemblies.
Don't use the same name for the namespace and type.
Don't pluralize company and product names and acronyms.

[ 64 ]
Classes, Objects, and Data Structures Chapter 3

We'll move on to the responsibility of classes.

A class should have only one responsibility

Responsibility is the work that has been assigned to the class. In the SOLID set of
principles, the S stands for Single Responsibility Principle (SRP). When applied to a class,
SRP states that the class must only work on a single aspect of the feature being
implemented. The responsibility of that single aspect should be fully encapsulated within
the class. Therefore, you should never apply more than one responsibility to a class.

Let's look at an example to understand why:

public class MultipleResponsibilities()
{
public string DecryptString(string text,
SecurityAlgorithm algorithm)
{
// ...implementation...
}

public string EncryptString(string text,

SecurityAlgorithm algorithm)
{
// ...implementation...
}

public string ReadTextFromFile(string filename)

{
// ...implementation...
}

public string SaveTextToFile(string text, string filename)

{
// ...implementation...
}
}

As you can see in the preceding code, for the MultipleResponsibilities class, we have
our cryptography functionalities implemented with the DecryptString and the
EncryptString methods. We also have file access implemented with the
ReadTextFromFile and SaveTextToFile methods. This class breaks the SRP principle.

[ 65 ]
Classes, Objects, and Data Structures Chapter 3

So we need to break this class up into two classes, one for cryptography and the other for
file access:
namespace FakeCompany.Core.Security
{
public class Cryptography
{
public string DecryptString(string text,
SecurityAlgorithm algorithm)
{
// ...implementation...
}

public string EncryptString(string text,

SecurityAlgorithm algorithm)
{
// ...implementation...
}
}
}

As we can now see from the preceding code, by moving the EncryptString and
DecryptString methods to their own Cryptography class in the core security
namespace, we have made it easy to reuse the code to encrypt and decrypt strings across
different products and technology groups. The Cryptography class also complies with
SRP.

In the following code, we can see that the SecurityAlgorithm parameter of the
Cryptography class is an enum and has been placed in its own source file. This helps to
keep code clean, minimal, and well organized:
using System;

namespace FakeCompany.Core.Security
{
[Flags]
public enum SecurityAlgorithm
{
Aes,
AesCng,
MD5,
SHA5
}
}

[ 66 ]
Classes, Objects, and Data Structures Chapter 3

Now, in the following TextFile class, we again abide by SRP and have a nice reusable
class that is in the appropriate core filesystem namespace. The TextFile class is reusable
across different products and technology groups:
namespace FakeCompany.Core.FileSystem
{
public class TextFile
{
public string ReadTextFromFile(string filename)
{
// ...implementation...
}

public string SaveTextToFile(string text, string filename)

{
// ...implementation...
}
}
}

We've looked at the organization and the responsibility of classes. Now let's take a look at
commenting on classes for the benefit of other developers.

Commenting for documentation generation

Documenting your source code is always a good idea, whether it is an internal project or
external software that will be used by other developers. Internal projects suffer because of
developer turnover and often poor, or little to no documentation available to help new
developers get up to speed. Many third-party APIs fail to get off the ground or uptake is
slower than expected, often with adopters abandoning the APIs through frustration
because of the poor state of the developer documentation.

It is always a good idea to include copyright notices at the top of each source code file and
to comment on your namespaces, interfaces, classes, enums, structs, methods, and
properties. Your copyright comments should be first in the source file, above the using
statements and take the form of a multiline comment that starts with /* and ends with */:
/**************************************************************************
********
* Copyright 2019 PacktPub
*
* Permission is hereby granted, free of charge, to any person obtaining a
copy of
* this software and associated documentation files (the "Software"), to

[ 67 ]
Classes, Objects, and Data Structures Chapter 3

deal in
* the Software without restriction, including without limitation the
rights to use,
* copy, modify, merge, publish, distribute, sublicense, and/or sell copies
of the
* Software, and to permit persons to whom the Software is furnished to do
so,
* subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included
in all
* copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
FROM,
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
IN THE
* SOFTWARE.
***************************************************************************
******/

using System;

/// <summary>
/// The CH3.Core.Security namespace contains fundamental types used
/// for the purpose of implementing application security.
/// </summary>
namespace CH3.Core.Security
{
/// <summary>
/// Encrypts and decrypts provided strings based on the selected
/// algorithm.
/// </summary>
public class Cryptography
{
/// <summary>
/// Decrypts a string using the selected algorithm.
/// </summary>
/// <param name="text">The string to be decrypted.</param>
/// <param name="algorithm">
/// The cryptographic algorithm used to decrypt the string.
/// </param>

[ 68 ]
Classes, Objects, and Data Structures Chapter 3

/// <returns>Decrypted string</returns>

public string DecryptString(string text,
SecurityAlgorithm algorithm)
{
// ...implementation...
throw new NotImplementedException();
}

/// <summary>
/// Encrypts a string using the selected algorithm.
/// </summary>
/// <param name="text">The string to encrypt.</param>
/// <param name="algorithm">
/// The cryptographic algorithm used to encrypt the string.
/// </param>
/// <returns>Encrypted string</returns>
public string EncryptString(string text,
SecurityAlgorithm algorithm)
{
// ...implementation...
throw new NotImplementedException();
}
}
}

The preceding code sample provides an example of a documented namespace and class
with documented methods. You will see that the documentation comments for the
namespace and contained members start with the documentation comment /// and are
directly above the item being commented on. When you type the three forward slashes,
Visual Studio automatically generates the XML tags based on the line below.

For example, in the preceding code, the namespace only has a summary and so does the
class, but both methods contain a summary, a couple of parameter comments, and a return
comment.

The following table contains the different XML tags that you can use in your
documentation comments.

Tag Section Purpose

<c> <c> Formats text as code
<code> <code> Provides source code as output
<example> <example> Provides an example
<exception> <exception> Describes the exceptions that can be thrown by the method
<include> <include> Includes XML from an external file
<list> <list> Adds a list or table

[ 69 ]
Classes, Objects, and Data Structures Chapter 3

<para> <para> Adds structure to text

<param> <param> Describes the parameter of a constructor or method
<paramref> <paramref> Tags a word to identify it is a parameter
<permission> <permission> Describes the security accessibility of the member
<remarks> <remarks> Provides additional information
<returns> <returns> Describes the return type
<see> <see> Adds a hyperlink
<seealso> <seealso> Adds a see also entry
<summary> <summary> Summarizes the type or member
<value> <value> Describes the value
<typeparam> Describes the type parameter
<typeparamref> Tags a word to identify it as a type parameter

From the preceding table, it is clear that you have plenty of scope for documenting your
source code. So it is a good idea to make the best use of the available tags to document your
code. The better the documentation, the quicker and easier it will be for other developers to
get up to speed with using the code.

It is now time to look at cohesion and coupling.

Cohesion and coupling

In a well-designed C# assembly, code will be correctly grouped together. This is known as
high cohesion. Low cohesion is when you have code grouped together that does not
belong together.

You want related classes to be as independent as possible. The more dependent one class is
on another class, the higher the coupling. This is known as tight coupling. The more
independent classes are of one another, the lower the cohesion. This is known as low
cohesion.

So, in a well-defined class, you want high cohesion and low coupling. We'll now look at
examples of tight coupling followed by low coupling.

[ 70 ]
Classes, Objects, and Data Structures Chapter 3

An example of tight coupling

In the following code example, the TightCouplingA class breaks encapsulation and makes
the _name variable directly accessible. The _name variable should be private and modified
only by the properties of methods within its enclosing class. The Name property provides
get and set methods to validate the _name variable, but this is pretty pointless as those
checks can be bypassed and the properties not called:
using System.Diagnostics;

namespace CH3.Coupling
{
public class TightCouplingA
{
public string _name;

public string Name

{
get
{
if (!_name.Equals(string.Empty))
return _name;
else
return "String is empty!";
}
set
{
if (value.Equals(string.Empty))
Debug.WriteLine("String is empty!");
}
}
}
}

On the other hand, in the following code, the TightCouplingB class creates an instance of
TightCouplingA. It then introduces tight coupling between the two classes by directly
accessing the _name member variable and setting it to null, and then directly accessing to
print its value to the debug output window:
using System.Diagnostics;

namespace CH3.Coupling
{
public class TightCouplingB
{
public TightCouplingB()

[ 71 ]
Classes, Objects, and Data Structures Chapter 3

{
TightCouplingA tca = new TightCouplingA();
tca._name = null;
Debug.WriteLine("Name is " + tca._name);
}
}
}

Now let's look at the same simple example using a low coupling.

An example of low coupling

In this example, we have two classes, LooseCouplingA and LooseCouplingB.
LooseCouplingA declares a private instance variable named _name, and this variable is set
via a public property.

LooseCouplingB creates an instance of LooseCouplingA and gets and sets the value of
Name. Because the _name data member cannot be set directly, the checks on setting and
getting the value of that data member are performed.

And so we have an example of loose coupling. Let's have a look at the two classes called
LooseCouplingA and LooseCouplingB that show this in action:

using System.Diagnostics;

namespace CH3.Coupling
{
public class LooseCouplingA
{
private string _name;
private readonly string _stringIsEmpty = "String is empty";

public string Name

{
get
{
if (_name.Equals(string.Empty))
return _stringIsEmpty;
else
return _name;
}

set
{
if (value.Equals(string.Empty))

[ 72 ]
Classes, Objects, and Data Structures Chapter 3

Debug.WriteLine("Exception: String length must be

greater than zero.");
}
}
}
}

In the LooseCouplingA class, we declare the _name field as private and so prevent the
data from being directly modified. The _name data is made indirectly accessible by the
Name property:

using System.Diagnostics;

namespace CH3.Coupling
{
public class LooseCouplingB
{
public LooseCouplingB()
{
LooseCouplingA lca = new LooseCouplingA();
lca = null;
Debug.WriteLine($"Name is {lca.Name}");
}
}
}

The LooseCouplingB class is unable to directly access the _name variable of the
LooseCouplingB class, and so modifies the data member via a property.

Well, we've looked at coupling and now know how to avoid tightly coupled code and
implement loosely coupled code. So now, it is time for us to look at some examples of low
cohesion and high cohesion.

An example of low cohesion

When a class has more than one responsibility, it is said to be a low cohesive class. Have a
look at the following code:
namespace CH3.Cohesion
{
public class LowCohesion
{
public void ConnectToDatasource() { }
public void ExtractDataFromDataSource() { }
public void TransformDataForReport() { }

[ 73 ]
Classes, Objects, and Data Structures Chapter 3

public void AssignDataAndGenerateReport() { }

public void PrintReport() { }
public void CloseConnectionToDataSource() { }
}
}

As we can see, the preceding class has at least three responsibilities:

Connecting to and disconnecting from a data source

Extracting data and transforming it ready for report insertion
Generating a report and printing it out

You will see clearly how this breaks the SRP. Next, we will break this class down into three
classes that adhere to the SRP.

An example of high cohesion

In this example, we are going to break down the LowCohesion class into three classes that
obey the SRP. These will be called Connection, DataProcessor, and ReportGenerator.
Let's see how much cleaner the code is after we implement the three classes.

In the following class, you can see that the only methods in that class are related to
connecting to a data source:
namespace CH3.Cohesion
{
public class Connection
{
public void ConnectToDatasource() { }
public void CloseConnectionToDataSource() { }
}
}

The class itself is named Connection, so this is an example of a high cohesive class.

In the following code, the DataProcessor class contains two methods that process data by
extracting data from the data source and transforming that data for insertion into the
report:
namespace CH3.Cohesion
{
public class DataProcessor
{
public void ExtractDataFromDataSource() { }
public void TransformDataForReport() { }

[ 74 ]
Classes, Objects, and Data Structures Chapter 3

}
}

So this is another example of a highly cohesive class.

In the following code, the ReportGenerator class only has methods associated with
generating and outputting the report:
namespace CH3.Cohesion
{
public class ReportGenerator
{
public void AssignDataAndGenerateReport() { }
public void PrintReport() { }
}
}

Again, this is another example of a highly cohesive class.

Looking at each of the three classes, we can see that they contain only methods that pertain
to their single responsibility. And so each of the three preceding classes is highly cohesive.

It is now time to look at how we design our code for change by using interfaces in place of
classes so that code can be injected into constructors and methods using dependency
injection and inversion of control.

Design for change

When designing for change, you should change the what to the how.

The what is the requirement of the business. As any seasoned person involved in a role
within software development will tell you that requirements frequently change. As such,
the software has to be adaptable to meet those changes. The business is not interested in
how the requirements are implemented by the software and infrastructure teams, only that
the requirements are met precisely on time and on budget.

On the other hand, the software and infrastructure teams are more focused on how those
business requirements are to be met. Regardless of the technology and processes that are
adopted for the project to implement the requirements, the software and target
environment must be adaptable to changing requirements.

[ 75 ]
Classes, Objects, and Data Structures Chapter 3

But that is not all. You see, software versions often change with bug fixes and new features.
As new features are implemented and refactoring takes place, the software code becomes
deprecated and eventually obsolete. On top of that, software vendors have a road map of
their software that forms part of their application life cycle management. Eventually,
software versions get to the point where they are retired and no longer supported by the
vendor. This can force a major migration from the current version, which will no longer be
supported, to the new supported version, and this can bring with it breaking changes that
must be addressed.

Interface-oriented programming
Interface-Oriented Programming (IOP) helps us to program polymorphic code.
Polymorphism in OOP is defined as different classes having their own implementations of
the same interface. And so, by using interfaces, we can morph our software to meet the
needs of the business.

Let's consider a database connection example. An application may be required to connect to

different data sources. But how can the database code remain the same no matter what
database is employed? Well, the answer lies in the use of interfaces.

You have different database connection classes that implement the same database
connection interface, but they each have their own versions of the implemented methods.
This is known as polymorphism. The database then accepts a database connection
parameter that is of the database connection interface type. You can then pass into the
database any database connection type that implements the database connection interface.
Let's code this example so that it makes things a little more clear.

Start by creating a simple .NET Framework console application. Then update the Program
class as follows:
static void Main(string[] args)
{
var program = new Program();
program.InterfaceOrientedProgrammingExample();
}

private void InterfaceOrientedProgrammingExample()

{
var mongoDb = new MongoDbConnection();
var sqlServer = new SqlServerConnection();
var db = new Database(mongoDb);
db.OpenConnection();
db.CloseConnection();

[ 76 ]
Classes, Objects, and Data Structures Chapter 3

db = new Database(sqlServer);
db.OpenConnection();
db.CloseConnection();
}

In this code, the Main() method creates a new instance of the Program class and then calls
the InterfaceOrientedProgrammingExample() method. In that method, we instantiate
two different database connections, one for MongoDB and one for SQL Server. We then
instantiate the database with a MongoDB connection, open the database connection, and
then close it. Then we instantiate a new database using the same variable and pass in a SQL
Server connection, then open the connection and close the connection. As you can see, we
only have one Database class with a single constructor, yet the Database class will work
with any database connection that implements the required interface. So, let's add the
IConnection interface:

public interface IConnection

{
void Open();
void Close();
}

The interface has only two methods called Open() and Close(). Add the MongoDB class
that will implement this interface:
public class MongoDbConnection : IConnection
{
public void Close()
{
Console.WriteLine("Closed MongoDB connection.");
}

public void Open()

{
Console.WriteLine("Opened MongoDB connection.");
}
}

We can see that the class implements the IConnection interface. Each method prints out a
message to the console. Now add that SQLServerConnection class:
public class SqlServerConnection : IConnection
{
public void Close()
{
Console.WriteLine("Closed SQL Server Connection.");
}

[ 77 ]
Classes, Objects, and Data Structures Chapter 3

public void Open()

{
Console.WriteLine("Opened SQL Server Connection.");
}
}

The same goes for the Database class. It implements the IConnection interface, and for
each method invocation, a message is printed to the console. And now for the Database
class, as follows:
public class Database
{
private readonly IConnection _connection;

public Database(IConnection connection)

{
_connection = connection;
}

public void OpenConnection()

{
_connection.Open();
}

public void CloseConnection()

{
_connection.Close();
}
}

The Database class accepts an IConnection parameter. This sets the _connection
member variable. The OpenConnection() method opens the database connection, and the
CloseConnection() method closes the database connection. Well, it's time to run the
program. You should see the following output in the console window:
Opened MongoDB connection.
Closed MongoDB connection.
Opened SQL Server Connection.
Closed SQL Server Connection.

So now, you can see the advantage of programming to interfaces. You can see how they
enable us to extend the program without having to modify the existing code. That means
that if we need to support more databases, then all we have to do is write more connection
objects that implement the IConnection interface.

[ 78 ]
Classes, Objects, and Data Structures Chapter 3

Now that you know how interfaces work, we can look at how to apply them to dependency
injection and inversion of control. Dependency injection helps us to write clean code that is
loosely coupled and easy to test, and inversion of control enables the interchanging of
software implementations as required, as long as those implementations implement the
same interface.

Dependency injection and inversion of control

In C#, we have the ability to address changing software needs using Dependency
Injection (DI) and Inversion of Control (IoC). These two terms do have different meanings
but are often used interchangeably to mean the same thing.

With IoC, you program a framework that accomplishes tasks by calling modules. An IoC
container is used to keep a register of modules. These modules are loaded when requested
by the user or configuration requests them.

DI removes internal dependencies from classes. Dependent objects are then injected by an
external caller. An IoC container uses DI to inject dependent objects into an object or
method.

In this chapter, you will find some useful resources that will help you to understand IoC
and DI. You will then be able to use these techniques in your programs.

Let's see how we can implement our own simple DI and IoC without any third-party
frameworks.

An example of DI
In this example, we are going to roll our own simple DI. We will have an ILogger interface
that will have a single method with a string parameter. We will then produce a class called
TextFileLogger that implements the ILogger interface and outputs a string to a text file.
Finally, we will have a Worker class that will demonstrate constructor injection and
method injection. Let's look at the code.

The following interface has a single method that will be used for implementing classes to
output a message according to the implementation of the method:
namespace CH3.DependencyInjection
{
public interface ILogger
{
void OutputMessage(string message);

[ 79 ]
Classes, Objects, and Data Structures Chapter 3

}
}

The TexFileLogger class implements the ILogger interface and outputs the message to a
text file:
using System;

namespace CH3.DependencyInjection
{
public class TextFileLogger : ILogger
{
public void OutputMessage(string message)
{
System.IO.File.WriteAllText(FileName(), message);
}

private string FileName()

{
var timestamp = DateTime.Now.ToFileTimeUtc().ToString();
var path = Environment.GetFolderPath(Environment
.SpecialFolder.MyDocuments);
return $"{path}_{timestamp}";
}
}
}

The Worker class provides an example of constructor DI and method DI. Notice that the
parameter is an interface. So, any class that implements that interface can be injected at
runtime:
namespace CH3.DependencyInjection
{
public class Worker
{
private ILogger _logger;

public Worker(ILogger logger)

{
_logger = logger;
_logger.OutputMessage("This constructor has been injected
with a logger!");
}

public void DoSomeWork(ILogger logger)

{
logger.OutputMessage("This methods has been injected
with a logger!");

[ 80 ]
Classes, Objects, and Data Structures Chapter 3

}
}
}

The DependencyInject method runs the example to show DI in action:

private void DependencyInject()
{
var logger = new TextFileLogger();
var di = new Worker(logger);
di.DoSomeWork(logger);
}

As you can see with the code we've just looked at, we start by producing a new instance of
the TextFileLogger class. This object is then injected into the constructor of the worker.
We then call the DoSomeWork method and pass in the TextFileLogger instance. In this
simple example, we have seen how to inject code into a class via its constructor and via
methods.

What is good about this code is it removes the dependency between the worker and the
TextFileLogger instance. This makes it easy for us to replace the
TextFileLogger instance with any other type of logger that implements the ILogger
interface. So we could have used, for example, an event viewer logger or even a database
logger. Using DI is a good way to reduce coupling in your code.

Now that we've seen DI at work, we should also look at IoC. And we'll do that now.

An example of IoC
In this example, we are going to register dependencies with an IoC container. We will then
use DI to inject the necessary dependencies.

In the following code, we have an IoC container. The container registers the dependencies
to be injected in a dictionary, and reads values from the configuration metadata:
using System;
using System.Collections.Generic;

namespace CH3.InversionOfControl
{
public class Container
{
public delegate object Creator(Container container);

private readonly Dictionary<string, object> configuration = new

[ 81 ]
Classes, Objects, and Data Structures Chapter 3

Dictionary<string, object>();
private readonly Dictionary<Type, Creator> typeToCreator = new
Dictionary<Type, Creator>();

public Dictionary<string, object> Configuration

{
get { return configuration; }
}

public void Register<T>(Creator creator)

{
typeToCreator.Add(typeof(T), creator);
}

public T Create<T>()
{
return (T)typeToCreator[typeof(T)](this);
}

public T GetConfiguration<T>(string name)

{
return (T)configuration[name];
}
}
}

Then, we create a container, and we use the container to configure metadata, register types,
and create instances of dependencies:
private void InversionOfControl()
{
Container container = new Container();
container.Configuration["message"] = "Hello World!";
container.Register<ILogger>(delegate
{
return new TextFileLogger();
});
container.Register<Worker>(delegate
{
return new Worker(container.Create<ILogger>());
});
}

Next up, we will look at how to limit an object's knowledge to knowing only about its close
relatives using the Law of Demeter. This will help us to write a clean C# code that avoids
the use of navigation trains.

[ 82 ]
Classes, Objects, and Data Structures Chapter 3

The Law of Demeter

The Law of Demeter aims to remove navigation trains (dot counting), and it also aims to
provide good encapsulation with loosely coupled code.

A method that understands a navigation train breaks the Law of Demeter. For example,
have a look at the following code:
report.Database.Connection.Open(); // Breaks the Law of Demeter.

Each unit of code should have a limited amount of knowledge. That knowledge should
only be of relevant code that is closely related. With the Law of Demeter, you must tell and
not ask. Using this law, you may only call methods of objects that are one or more of the
following:

Passed as arguments
Created locally
Instance variables
Globals

Implementing the Law of Demeter can be difficult, but there are advantages to telling
rather than asking. One such benefit is the decoupling of your code.

It is good to see a bad example that breaks the Law of Demeter, along with one that obeys
the Law of Demeter, so we will see this in the following sections.

A good and a bad example (chaining) of the Law

of Demeter
In the good example, we have the report instance variable. On the report variable object
instance, the method to open the connection is called. This does not break the law.

The following code is a Connection class with a method that opens a connection:
namespace CH3.LawOfDemeter
{
public class Connection
{
public void Open()
{
// ... implementation ...
}

[ 83 ]
Classes, Objects, and Data Structures Chapter 3

}
}

The Database class creates a new Connection object and opens a connection:
namespace CH3.LawOfDemeter
{
public class Database
{
public Database()
{
Connection = new Connection();
}

public Connection Connection { get; set; }

public void OpenConnection()

{
Connection.Open();
}
}
}

In the Report class, a Database object is instantiated and then a connection to the database
is opened:
namespace CH3.LawOfDemeter
{
public class Report
{
public Report()
{
Database = new Database();
}

public Database Database { get; set; }

public void OpenConnection()

{
Database.OpenConnection();
}
}
}

So far, we have seen good code that obeys the Law of Demeter. But the following is code
that breaks this law.

[ 84 ]
Classes, Objects, and Data Structures Chapter 3

In the Example class, the Law of Demeter is broken because we introduce method chaining,
as in report.Database.Connection.Open():
namespace CH3.LawOfDemeter
{
public class Example
{
public void BadExample_Chaining()
{
var report = new Report();
report.Database.Connection.Open();
}

public void GoodExample()

{
var report = new Report();
report.OpenConnection();
}
}
}

In this bad example, the Database getter is called on the report instance variable. This is
acceptable. But then a call is made to the Connection getter that returns a different object.
This breaks the Law of Demeter, as does the final call to open the connection.

Immutable objects and data structures

Immutable types are normally thought of as just value types. With value types, it makes
sense that when they are set, you don't want them to change. But you can also have
immutable object types and immutable data structure types. Immutable types are a type
whose internal state does not change once they have been initialized.

The behavior of immutable types does not astonish or surprise fellow programmers and so
conforms to the principle of least astonishment (POLA). The POLA conformity of
immutable types adheres to any contracts made between clients, and because it is
predictable, programmers will find it easy to reason about its behavior.

Since immutable types are predictable and do not change, you are not going to be in for any
nasty surprises. So you don't have to worry about any undesirable effects due to them
being altered in some way. This makes immutable types ideal for sharing between threads
as they are thread-safe and there is no need for defensive programming.

[ 85 ]
Classes, Objects, and Data Structures Chapter 3

When you create an immutable type and use object validation, you have a valid object for
the lifetime of that object.

Let's have a look at an example of an immutable type in C#.

An example of an immutable type

We are now going to look at an immutable object. The Person object in the following code
has three private member variables. The only time these can be set is during the creation
time in the constructor. Once set, they are unable to be modified for the rest of the object's
lifetime. Each variable is only readable via read-only properties:
namespace CH3.ImmutableObjectsAndDataStructures
{
public class Person
{
private readonly int _id;
private readonly string _firstName;
private readonly string _lastName;

public int Id => _id;

public string FirstName => _firstName;
public string LastName => _lastName;
public string FullName => $"{_firstName} {_lastName}";
public string FullNameReversed => $"{_lastName}, {_firstName}";

public Person(int id, string firstName, string lastName)

{
_id = id;
_firstName = firstName;
_lastName = lastName;
}
}
}

Now we have seen how easy it is to write immutable objects and data structures, we will
look at data and methods in objects.

[ 86 ]
Classes, Objects, and Data Structures Chapter 3

Objects should hide data and expose

methods
The state of your object is stored in member variables. These member variables are data.
Data should not be directly accessible. You should only provide access to data via exposed
methods and properties.

Why should you hide your data and expose your methods?

Hiding data and exposing methods is known in the OOP world as encapsulation.
Encapsulation hides the inner workings of a class from the outside world. This makes it
easy to be able to change value types without breaking existing implementations that rely
on the class. Data can be made read/writable, writable, or read-only providing more
flexibility to you regarding data access and usage. You can also validate input and so
prevent data from receiving invalid values. Encapsulating also makes testing your classes
much easier, and you can make your classes more reusable and extendable.

Let's look at an example.

An example of encapsulation
The following code example shows an encapsulated class. The Car object is mutable. It has
properties that get and set the data values once they have been initialized by the
constructor. The constructor and the set properties perform the validation of the parameter
arguments. If the value is invalid, an invalid argument exception is thrown, otherwise the
value is passed back and the data value is set:
using System;

namespace CH3.Encapsulation
{
public class Car
{
private string _make;
private string _model;
private int _year;

public Car(string make, string model, int year)

{
_make = ValidateMake(make);
_model = ValidateModel(model);
_year = ValidateYear(year);
}

[ 87 ]
Classes, Objects, and Data Structures Chapter 3

private string ValidateMake(string make)

{
if (make.Length >= 3)
return make;
throw new ArgumentException("Make must be three
characters or more.");
}

public string Make

{
get { return _make; }
set { _make = ValidateMake(value); }
}

// Other methods and properties omitted for brevity.

}
}

The benefit of the preceding code is that if you need to change the validation for the code
that gets or sets the data values, you can do so without breaking the implementation.

Data structures should expose data and

have no methods
Structures differ from classes in that they use value equality in place of reference equality.
Other than that, there is not much difference between a struct and a class.

There is a debate as to whether a data structure should make the variables public or hide
them behind get and set properties. It is purely down to you which you choose, but
personally I always think it best to hide data even in structs and only provide access via
properties and methods. There is one caveat in terms of having clean data structures that
are safe, and that is that once created, structs should not allow themselves to be mutated by
methods and get properties. The reason for this is that changes to temporary data structures
will be discarded.

Let's now look at a simple data structure example.

[ 88 ]
Classes, Objects, and Data Structures Chapter 3

An example of data structure

The following code is a simple data structure:
namespace CH3.Encapsulation
{
public struct Person
{
public int Id { get; set; }
public string FirstName { get; set; }
public string LastName { get; set; }

public Person(int id, string firstName, string lastName)

{
Id = id;
FirstName = firstName;
LastName = lastName;
}
}
}

As you can see, the data structure is not that much different from a class in that it has a
constructor and properties.

With this, we come to the end of the chapter and will now review what we've learned.

Summary
In this chapter, we learned about organizing our namespaces in folders and packages, and
how good organization can help to prevent namespace classes. We then moved on to
classes and responsibility and looked at why classes should only have one responsibility.
We also looked at cohesion and coupling and why it is important to have high cohesion
and low coupling.

Good documentation requires public members to be correctly commented on in

documentation tools, and we saw how to do this using XML comments. The importance of
why you should design for change was also discussed with basic examples of DI and IoC.

The Law of Demeter showed you how to not to talk to strangers, but only immediate
friends, and how to avoid chaining. And finally, we looked at objects and data structures
and what they should hide and what they should make public.

[ 89 ]
Classes, Objects, and Data Structures Chapter 3

In the next chapter, we will briefly cover functional programming in C# and how to write
clean methods that are small. We will also learn to avoid having more than two parameters
in our methods, as methods with many parameters can become unwieldy. Plus we will
learn to avoid duplication which can be a troublesome source of bugs when fixed in one
location, but still exist elsewhere in your code.

Questions
1. How can we organize our classes in C#?
2. How many responsibilities should a class have?
3. How do you comment on your code for document generators?
4. What does cohesion mean?
5. What does coupling mean?
6. Should cohesion be high or low?
7. Should coupling be tight or loose?
8. What mechanisms are available that help you design for change?
9. What is DI?
10. What is IoC?
11. Name one benefit of using immutable objects.
12. What should objects hide and show?
13. What should structures hide and show?

Further reading
For more detail in regard to understanding the different kinds of cohesion and
coupling, check out https://www.geeksforgeeks.org/software-engineering-
coupling-and-cohesion/.
Many tutorials on IoC can be found at https://www.tutorialsteacher.
com/ioc/.

[ 90 ]
4
Writing Clean Functions
Clean functions are methods that are small (they have two or fewer arguments) and avoid
duplication. The ideal method has no parameters and does not modify the program's state.
Small methods are less prone to exceptions, so you will be writing much more robust code
that benefits you in the long run as you will have fewer bugs to fix.

Functional programming is a software coding methodology that treats computations as the

mathematical evaluation of computations. This chapter will teach you the benefits of
treating computations as the evaluation of mathematical functions in order to void
changing an object's state.

Large methods (also known as functions) can be unwieldy to read and prone to errors, so
writing small methods has its advantages. Hence, we will look at how large methods can be
broken up into smaller methods. In this chapter, we will cover functional programming in
C# and how to write small, clean methods.

Constructors and methods with multiple parameters can become a real pain to work with,
so we will have to look for ways to work around and pass multiple parameters, as well
as how to avoid using more than two parameters. The main reason for reducing the
number of parameters we have is that they can become hard to read, be a source of
irritation to fellow programmers, and cause visual stress if there are enough of them. They
can also be a sign that the method is trying to do too much, or that you need to consider
refactoring your code.

In this chapter, we will cover the following topics:

Understanding functional programming

Keeping methods small
Avoiding duplication
Avoiding multiple parameters
Writing Clean Functions Chapter 4

By the time you have worked through this chapter, you will have the skills to do the
following:

Describe what functional programming is

Provide existing examples of functional programming in the C# programming
language
Write functional C# code
Avoid writing methods with more than two arguments
Write immutable data objects and structures
Keep your methods small
Write code that adheres to the Single Responsibility Principle (SRP)

Let's get started!

Understanding functional programming

The only thing that sets functional programming aside from other methods of
programming is that functions do not modify data or state. You will use functional
programming in scenarios such as deep learning, machine learning, and artificial
intelligence when it is necessary to perform different sets of operations on the same set of
data.

The LINQ syntax within .NET Framework is an example of functional programming. So, if
you are wondering what functional programming looks like, and if you have used LINQ
before, then you have been subjected to functional programming and should know what it
looks like.

Since functional programming is a deep subject and many books, courses, and videos exist
on this topic, we will only touch on the topic briefly in this chapter by looking at pure
functions and immutable data.

A pure function is restricted to only operating on the data that is passed into it. As a result,
the method is predictable and avoids producing side effects. This benefits programmers
because such methods are easier to reason about and test.

[ 92 ]
Writing Clean Functions Chapter 4

Once an immutable data object or data structure has been initialized, the contained data
values will not be modified. Because the data is only set and not modified, you can easily
reason about what the data is, how it is set, and what the outcome of any operation will be,
given the inputs. Immutable data is also easier to test as you know what your inputs are
and what outputs are expected. This makes writing test cases much easier as you don't
have so many things to consider, such as object state. The benefit of immutable objects and
structures is that they are thread-safe. Thread-safe objects and structures make for good
data transfer objects (DTOs) that can be passed between threads.

But structs can still be mutable if they contain reference types. One way around this would
be to make the reference type immutable. C# 7.2 added support for readonly struct and
ImmutableStruct. So, even if our structures contain reference types, we can now use
these new C# 7.2 constructs to make structures with reference types immutable.

Now, let's have a look at a pure function example. The only way to set the properties of an
object is via the constructor at construction time. The class is a Player class whose only job
is to hold the name of the player and their high score. A method is provided that updates
the player's high score:
public class Player
{
public string PlayerName { get; }
public long HighScore { get; }

public Player(string playerName, long highScore)

{
PlayerName = playerName;
HighScore = highScore;
}

Public Player UpdateHighScore(long highScore)

{
return new Player(PlayerName, highScore);
}

Notice that the UpdateHighScore method does not update the HighScore property.
Instead, it instantiates and returns a new Player class by passing in the
PlayerName variable, which is already set in the class, and highScore, which is the
method parameter. You have now seen a very simple example of how to program your
software without changing its state.

[ 93 ]
Writing Clean Functions Chapter 4

Functional programming is a very large subject and requires a mind shift

that can be very difficult for both procedural and object-oriented
programmers. Since it is outside the scope of this book (to delve deep into
the topic of functional programming), you are actively encouraged to
peruse the functional programming resources on offer from PacktPub for
yourself.

Packt has some very good books and videos that specialize in teaching the
top tiers of functional programming. You will find links to some Packt
functional programming resources at the end of this chapter, in the Further
reading section.

Before we move on, we will look at some LINQ examples since LINQ is an example of
functional programming in C#. It will be good to have an example dataset. The following
code builds a list of vendors and products. We'll start by writing the Product structure:
public struct Product
{
public string Vendor { get; }
public string ProductName { get; }
public Product(string vendor, string productName)
{
Vendor = vendor;
ProductName = productName;
}
}

Now that we have our struct, we will add some sample data inside the GetProducts()
method:
public static List<Product> GetProducts()
{
return new List<Products>
{
new Product("Microsoft", "Microsoft Office"),
new Product("Oracle", "Oracle Database"),
new Product("IBM", "IBM DB2 Express"),
new Product("IBM", "IBM DB2 Express"),
new Product("Microsoft", "SQL Server 2017 Express"),
new Product("Microsoft", "Visual Studio 2019 Community Edition"),
new Product("Oracle", "Oracle JDeveloper"),
new Product("Microsoft", "Azure"),
new Product("Microsoft", "Azure"),
new Product("Microsoft", "Azure Stack"),
new Product("Google", "Google Cloud Platform"),
new Product("Amazon", "Amazon Web Services")

[ 94 ]
Writing Clean Functions Chapter 4

};
}

Finally, we can start to use LINQ on our list. In the preceding example, we will get a
distinct list of products, ordered by the vendor's names, and print out the results:
class Program
{
static void Main(string[] args)
{
var vendors = (from p in GetProducts()
select p.Vendor)
.Distinct()
.OrderBy(x => x);
foreach(var vendor in vendors)
Console.WriteLine(vendor);
Console.ReadKey();
}
}

Here, we obtain a list of vendors by calling GetProducts() and selecting only the Vendor
column. Then, we filter the list so that it only includes a vendor once by calling the
Distinct() method. The list of vendors is then ordered alphabetically by calling
OrderBy(x => x), where x is the vendor's name. Upon obtaining the ordered list of
distinct vendors, we then loop through the list and print the vendor's name. Finally, we
wait for the user to press any key to exit the program.

One of the benefits of functional programming is that your methods are much smaller than
the methods in other types of programming. Next, we will take a look at why it is good to
keep methods small, as well as the techniques we can use, including functional
programming.

Keeping methods small

While programming clean and readable code, it is important to keep the methods small.
Preferably, in the C# world, it is best to keep methods under 10 lines long. The perfect length
is no more than 4 lines. A good way to keep methods small is to consider if you should be
trapping for errors or bubbling them further up the call stack. With defensive
programming, you can become a little too defensive, and this can add to the amount of
code you find yourself writing. Besides, methods that trap errors will be longer than
methods that don't.

[ 95 ]
Writing Clean Functions Chapter 4

Let's consider the following code, which can throw an ArgumentNullException:

public UpdateView(MyEntities context, DataItem dataItem)
{
InitializeComponent();
try
{
DataContext = this;
_dataItem = dataItem;
_context = context;
nameTextBox.Text = _dataItem.Name;
DescriptionTextBox.Text = _dataItem.Description;
}
catch (Exception ex)
{
Debug.WriteLine(ex);
throw;
}
}

In the preceding code, we can clearly see that there are two locations where an
ArgumentNullException may be raised. The first line of code to potentially raise an
ArgumentNullException is nameTextBox.Text = _dataItem.Name;; the second line
of code that may potentially raise the same exception is DescriptionTextBox.Text =
_dataItem.Description;. We can see that the exception handler catches the exception
when it occurs, writes it to the console, and then simply throws it back up the stack.

Notice that, from a human reading perspective, there are 8 lines of code that form the
try/catch block.

You can completely replace the try/catch exception handling with a single line of text by
writing your own argument validator. To explain this, we will provide an example.

Let's start by looking at the ArgumentValidator class. The purpose of this class is to
throw an ArgumentNullException with the name of the method that contains the null
argument:
using System;
namespace CH04.Validators
{
internal static class ArgumentValidator
{
public static void NotNull(
string name,
[ValidatedNotNull] object value
)

[ 96 ]
Writing Clean Functions Chapter 4

{
if (value == null)
throw new ArgumentNullException(name);
}
}

[AttributeUsage(
AttributeTargets.All,
Inherited = false,
AllowMultiple = true)
]
internal sealed class ValidatedNotNullAttribute : Attribute
{
}
}

Now that we have our null validation class, we can perform the new way of validating
parameters for null values in our methods. So, let's look at a simple example:
public ItemsUpdateView(
Entities context,
ItemsView itemView
)
{
InitializeComponent();
ArgumentValidator.NotNull("ItemsUpdateView", itemView);
// ### implementation omitted ###
}

As you can clearly see, we have replaced the whole of the try catch block with a one-
liner at the top of the method. When this validation detects a null argument, an
ArgumentNullException is thrown, preventing the code from continuing. This makes the
code much easier to read, and also helps with debugging.

Now, we'll look at formatting functions with indentation so that they are easy to read.

Indenting code
A very long method is hard to read and follow at the best of times, especially when you
have to scroll through the method many times to get to the bottom of it. But having to do
that with methods that are not properly formatted with the correct levels of indentation can
be a real nightmare.

[ 97 ]
Writing Clean Functions Chapter 4

If you ever encounter any method code that is poorly formatted, then make it your own
responsibility, as a professional coder, to tidy the code up before you do anything else. Any
code between braces is known as a code block. Code within a code block should be
indented by one level. Code blocks within code blocks should also be indented by one level,
as shown in the following example:
public Student Find(List<Student> list, int id)
{
Student r = null;foreach (var i in list)
{
if (i.Id == id)
r = i; } return r;
}

The preceding example demonstrates bad indentation and also bad loop programming.
Here, you can see that a list of students is being searched in order to find and return a
student with the specified ID that was passed in as a parameter. What annoys some
programmers and reduces the performance of the application is that the loop in the
preceding code continues, even when the student has been found. We can improve the
indentation and the performance of the preceding code as follows:
public Student Find(List<Student> list, int id)
{
Student r = null;
foreach (var i in list)
{
if (i.Id == id)
{
r = i;
break;
}
}
return r;
}

In the preceding code, we have improved the formatting and made sure that the code is
properly indented. We've added a break to the for loop so that the foreach loop is
terminated when a match is found.

Not only is the code now more readable, but it also performs much better. Imagine that the
code is being run against a university with 73,000 students on campus and via distance
learning. Consider that if the student matches the ID is the first in the list, then without the
break statement, the code would have to run 72,999 unnecessary computations. You can
see how much of a difference the break statement makes to the performance of the
preceding code.

[ 98 ]
Writing Clean Functions Chapter 4

We have left the return value in its original location as the compiler can complain that not
all code paths return a value. This is also why we added the break statement. It is clear that
proper indentation improves the readability of the code, thus aiding the programmer's
understanding of it. This enables the programmer to make any changes that they deem
necessary.

Avoiding duplication
Code can be either DRY or WET. WET code stands for Write Every Time and is the
opposite of DRY, which stands for Don't Repeat Yourself. The problem with WET code is
that it is the perfect candidate for bugs. Let's say your test team or a customer finds a bug
and reports it to you. You fix the bug and pass it on, only for it to come back and bite you as
many times as that code is encountered within your computer program.

Now, we DRY our WET code by removing duplication. One way we can do this is by
extracting the code and putting it into a method and then centralizing the method in such a
way that it is accessible to all the areas of the computer program that need it.

Time for an example. Imagine that you have a collection of expense items that consist of
Name and Amount properties. Now, consider having to get the decimal Amount for an
expense item by Name.

Say you had to do this 100 times. For this, you could write the following code:
var amount = ViewModel
.ExpenseLines
.Where(e => e.Name.Equals("Life Insurance"))
.FirstOrDefault()
.Amount;

There is no reason why you can't write that same code 100 times. But there is a way to write
it only once, thus reducing the size of your codebase and making you more productive.
Let's have a look at how we can do this:
public decimal GetValueByName(string name)
{
return ViewModel
.ExpenseLines
.Where(e => e.Name.Equals(name))
.FirstOrDefault()
.Amount;
}

[ 99 ]
Writing Clean Functions Chapter 4

To extract the required value from the ExpenseLines collection within your ViewModel,
all you have to do is pass the name of the value you require into the
GetValueName(string name) method, as shown in the following code:

var amount = GetValueByName("Life Insurance");

That one line of code is very readable, and the lines of code to get the value are contained in
a single method. So, if the method needs to be changed for whatever reason (such as a bug
fix), you only have to modify the code in one place.

The next logical step to writing good functions is to have as few parameters as possible. In
the next section, we'll look at why we should have no more than two parameters, as well as
how to work with just parameters, even if we need plenty more.

Avoiding multiple parameters

Niladic methods are the ideal type of methods in C#. Such methods have no parameters
(also known as arguments). Monadic methods only have one parameter. Dyadic methods
have two parameters. Triadic methods have three parameters. Methods that have more
than three parameters are known as polyadic methods. You should aim to keep the number
of parameters to a minimum (preferably less than three).

In the ideal world of C# programming, you should do your best to avoid triadic and
polyadic methods. The reason for this is not because it is bad programming, but because it
makes your code easier to read and understand. Methods with lots of parameters can cause
visual stress to programmers, and can also be a source of irritation. IntelliSense can also be
difficult to read and understand as you add more parameters.

Let's look at a bad example of a polyadic method that updates a user's account information:
public void UpdateUserInfo(int id, string username, string firstName,
string lastName, string addressLine1, string addressLine2, string
addressLine3, string addressLine3, string addressLine4, string city, string
postcode, string region, string country, string homePhone, string
workPhone, string mobilePhone, string personalEmail, string workEmail,
string notes)
{
// ### implementation omitted ###
}

[ 100 ]
Writing Clean Functions Chapter 4

As shown by the UpdateUserInfo method, the code is horrible to read. How can we
modify the method so that it transforms from a polyadic method into a monadic method?
The answer is simple – we pass in a UserInfo object. First of all, before we modify the
method, let's take a look at our UserInfo class:
public class UserInfo
{
public int Id { get;set; }
public string Username { get; set; }
public string FirstName { get; set; }
public string LastName { get; set; }
public string AddressLine1 { get; set; }
public string AddressLine2 { get; set; }
public string AddressLine3 { get; set; }
public string AddressLine4 { get; set; }
public string City { get; set; }
public string Region { get; set; }
public string Country { get; set; }
public string HomePhone { get; set; }
public string WorkPhone { get; set; }
public string MobilePhone { get; set; }
public string PersonalEmail { get; set; }
public string WorkEmail { get; set; }
public string Notes { get; set; }
}

We now have a class that contains all the information we need to pass into the
UpdateUserInfo method. The UpdateUserInfo method can now be transformed from a
polyadic method into a monadic method, as follows:
public void UpdateUserInfo(UserInfo userInfo)
{
// ### implementation omitted ###
}

How much better does the preceding code look? It is smaller and much more readable. The
rule of thumb should be to have less than three parameters, and ideally none. If your class
is obeying the SRP, then consider implementing the parameter object pattern, as we have
done here.

[ 101 ]
Writing Clean Functions Chapter 4

Implementing SRP
All objects and methods that you write should, at most, have one responsibility and no
more. Objects can have multiple methods, but those methods, when combined, should all
work toward the single purpose of the object they belong to. Methods can call multiple
methods, where each does different things. But the method itself should only do one thing.

A method that knows and does far too much is known as a God method. And likewise, an
object that knows and does too much is known as a God object. God objects and methods
are hard to read, maintain, and debug. Such objects and methods can often have the same
bug repeated many times. People who are good at the programming craft will avoid God
objects and God methods. Let's look at a method that is doing more than one thing:
public void SrpBrokenMethod(string folder, string filename, string text,
emailFrom, password, emailTo, subject, message, mediaType)
{
var file = $"{folder}{filename}";
File.WriteAllText(file, text);
MailMessage message = new MailMessage();
SmtpClient smtp = new SmtpClient();
message.From = new MailAddress(emailFrom);
message.To.Add(new MailAddress(emailTo));
message.Subject = subject;
message.IsBodyHtml = true;
message.Body = message;
Attachment emailAttachment = new Attachment(file);
emailAttachment.ContentDisposition.Inline = false;
emailAttachment.ContentDisposition.DispositionType =
DispositionTypeNames.Attachment;
emailAttachment.ContentType.MediaType = mediaType;
emailAttachment.ContentType.Name = Path.GetFileName(filename);
message.Attachments.Add(emailAttachment);
smtp.Port = 587;
smtp.Host = "smtp.gmail.com";
smtp.EnableSsl = true;
smtp.UseDefaultCredentials = false;
smtp.Credentials = new NetworkCredential(emailFrom, password);
smtp.DeliveryMethod = SmtpDeliveryMethod.Network;
smtp.Send(message);
}

SrpBrokenMethod is clearly doing more than one thing, so it breaks the SRP. We will now
break this method down into a number of smaller methods that only do one thing. We will
also address the issue of the polyadic nature of the method in that it has more than two
parameters.

[ 102 ]
Writing Clean Functions Chapter 4

Before we begin to break down the method into smaller methods that do only one thing, we
need to look at all the actions that the method is performing. The method starts by writing
text to a file. It then creates an email message, assigns an attachment, and finally sends the
email. So, for this, we need methods for the following:

Write text to file

Create an email message
Add an email attachment
Send email

Looking at the current method, we have four parameters that are passed into it for writing
text to a file: one for the folder, one for the filename, one for the text, and one for the media
type. The folder and filename can be combined into a single parameter called filename. If
filename and folder are two separate variables that are used inside the calling code, then
they can be passed into the method as a single interpolated string, such as
$"{folder}{filename}".

As for the media type, this could be privately set inside a struct during construction time.
We could use that struct to set the properties we need so that we can pass the struct in with
the three properties as a single parameter. Let's look at the code that accomplishes this:
public struct TextFileData
{
public string FileName { get; private set; }
public string Text { get; private set; }
public MimeType MimeType { get; }

public TextFileData(string filename, string text)

{
Text = text;
MimeType = MimeType.TextPlain;
FileName = $"{filename}-{GetFileTimestamp()}";
}

public void SaveTextFile()

{
File.WriteAllText(FileName, Text);
}

private static string GetFileTimestamp()

{
var year = DateTime.Now.Year;
var month = DateTime.Now.Month;
var day = DateTime.Now.Day;
var hour = DateTime.Now.Hour;

[ 103 ]
Writing Clean Functions Chapter 4

var minutes = DateTime.Now.Minute;

var seconds = DateTime.Now.Second;
var milliseconds = DateTime.Now.Millisecond;
return
$"{year}{month}{day}@{hour}{minutes}{seconds}{milliseconds}";
}
}

The TextFileData constructor ensures that the FileName value is unique by calling the
GetFileTimestamp() method and appending it to the end of FileName. To save the text
file, we call the SaveTextFile() method. Notice that MimeType is set internally and is set
to MimeType.TextPlain. We could have simply hardcoded MimeType as MimeType =
"text/plain";, but the advantage of using an enum is that the code is reusable, with the
added benefit of you not having to remember the text for a specific MimeType or look it up
on the internet. Now, we'll code enum and add a description to the enum value:
[Flags]
public enum MimeType
{
[Description("text/plain")]
TextPlain
}

Well, we've got our enum, but now we need a way to extract the description so that it can be
easily assigned to a variable. Therefore, we will create an extension class that will enable us
to get the description of an enum. This enables us to set MimeType, as follows:
MimeType = MimeType.TextPlain;

Without the extension method, the value of MimeType would be 0. But with the extension
method, the value of MimeType is "text/plain". You can now reuse this extension in
other projects and build it up as you require.

The next class we will write is the Smtp class, whose responsibility is to send an email via
the Smtp protocol:
public class Smtp
{
private readonly SmtpClient _smtp;

public Smtp(Credential credential)

{
_smtp = new SmtpClient
{
Port = 587,
Host = "smtp.gmail.com",

[ 104 ]
Writing Clean Functions Chapter 4

EnableSsl = true,
UseDefaultCredentials = false,
Credentials = new NetworkCredential(
credential.EmailAddress, credential.Password),
DeliveryMethod = SmtpDeliveryMethod.Network
};
}
public void SendMessage(MailMessage mailMessage)
{
_smtp.Send(mailMessage);
}
}

The Smtp class has a constructor that takes a single parameter of the Credential type.
This credential is used to log into the email server. The server is configured in the
constructor. When the SendMessage(MailMessage mailMessage) method is called, the
message is sent.

Let's write a DemoWorker class that splits the work into different methods:
public class DemoWorker
{
TextFileData _textFileData;

public void DoWork()

{
SaveTextFile();
SendEmail();
}

public void SendEmail()

{
Smtp smtp = new Smtp(new Credential("[email protected]",
"fakeP@55w0rd"));
smtp.SendMessage(GetMailMessage());
}

private MailMessage GetMailMessage()

{
var msg = new MailMessage();
msg.From = new MailAddress("[email protected]");
msg.To.Add(new MailAddress("[email protected]"));
msg.Subject = "Some subject";
msg.IsBodyHtml = true;
msg.Body = "Hello World!";
msg.Attachments.Add(GetAttachment());
return msg;
}

[ 105 ]
Writing Clean Functions Chapter 4

private Attachment GetAttachment()

{
var attachment = new Attachment(_textFileData.FileName);
attachment.ContentDisposition.Inline = false;
attachment.ContentDisposition.DispositionType =
DispositionTypeNames.Attachment;
attachment.ContentType.MediaType =
MimeType.TextPlain.Description();
attachment.ContentType.Name =
Path.GetFileName(_textFileData.FileName);
return attachment;
}
private void SaveTextFile()
{
_textFileData = new TextFileData(
$"{Environment.SpecialFolder.MyDocuments}attachment",
"Here is some demo text!"
);
_textFileData.SaveTextFile();
}
}

The DemoWorker class shows a much cleaner version of sending an email message. The
main method responsible for saving an attachment and sending it as an attachment via
email is called DoWork(). This method only contains two lines of code. The first line calls
the SaveTextFile() method, while the second line calls the SendEmail() method.

The SaveTextFile() method creates a new TextFileData struct and passes in the
filename and some text. It then calls the SaveTextFile() method in the TextFileData
struct, which is responsible for saving the text to the file specified.

The SendEmail() method creates a new Smtp class. The Smtp class has a Credential
parameter, while the Credential class has two string parameters for email address and
password. The email and password are used to log into the SMTP server. Once the SMTP
server has been created, the SendMessage(MailMessage mailMessage) method is
called.

This method requires a MailMessage object to be passed in. So, we have a method called
GetMailMethod() that builds a MailMessage object that is then passed into the
SendMessage(MailMessage mailMessage) method. GetMailMethod() adds an
attachment to MailMessage by calling the GetAttachment() method.

[ 106 ]
Writing Clean Functions Chapter 4

As you can see from these modifications, our code is now more compact and readable. That
is the key to good quality code that is easy to modify and maintain: it must be easy to read
and understand. That is why it is important for your methods to be small and clean with as
few parameters as possible.

Does your method break the SRP? If it does, you should consider breaking the method up
into as many methods as there are responsibilities. And that concludes this chapter on
writing clean functions. It is now time to summarize what you have learned and test your
knowledge.

Summary
In this chapter, you have seen how functional programming can improve the safety of your
code by not modifying the state, which can give rise to bugs, especially in multithreaded
applications. By keeping methods small with meaningful names and no more than two
parameters, you have seen how much cleaner your code is and easier to read. You have
also seen how we can remove duplication in our code and the benefits of doing so. Code
that is easy to read is easier to maintain and extend than code that is hard to read and
decipher!

We will now move on and look at the topic of exception handling. In the next chapter, you
will learn how to use exception handling appropriately, write your own custom C#
exceptions that provide meaningful information, and write code that avoids raising
NullPointerExceptions.

Questions
1. What do you call a method that has no parameters?
2. What do you call a method that has one parameter?
3. What do you call a method that has two parameters?
4. What do you call a method that has three parameters?
5. What do you call a method that has more than three parameters?
6. What two method types should be avoided and why?
7. In layman's terms, what is functional programming?
8. What are some advantages of functional programming?
9. Name one disadvantage of functional programming.
10. What is WET code, and why should it be avoided?

[ 107 ]
Writing Clean Functions Chapter 4

11. What is DRY code, and why should you use it?
12. How do you DRY out WET code?
13. Why should methods be as small as possible?
14. How do you implement validation without having to implement try/catch
blocks?

Further reading
Here are some additional resources so that you can delve deeper into the realms of C#
functional programming:

Functional C# by Wisnu Anggoro: https://www.packtpub.com/application-

development/functional-c. This book is devoted to C# functional programming
and is a good place to start if you want to know more.
Functional Programming in C# by Jovan Poppavic (MSFT): https://www.
codeproject.com/Articles/375166/Functional-programming-in-Csharp. This
is an in-depth article on functional C# programming. It contains diagrams and
has a 5-star rating.

[ 108 ]
5
Exception Handling
In the previous chapter, we looked at functions. Despite the best efforts of programmers to
write robust code, functions will, at some point, generate exceptions. This could be for a
number of reasons, such as a missing file or folder, an empty or null value, the location
can't be written to, or the user is denied access. So, with that in mind, in this chapter, you
will learn about appropriate ways to use exception handling to produce clean C# code.
First, we will start by looking at checked and unchecked exceptions with regards to
arithmetic OverflowExceptions. We will look at what they are, why they are used, and
some examples of them being used in code.

Then, we'll look at how we can avoid the NullPointerReference exception. After that,
we'll look at implementing specific business rules for specific types of exceptions. With our
fresh understanding of exceptions and exception business rules, we will set about building
our own custom exceptions and then finish off by looking at why we should not use
exceptions to control the flow of our computer programs.

In this chapter, we will cover the following topics:

Checked and unchecked exceptions

Avoiding NullPointerExceptions
Business rule exceptions
Exceptions should provide meaningful information
Building your own custom exceptions

By the end of this chapter, you will have the skills to do the following:

You will be able to understand what checked and unchecked exceptions are, and
why they are in C#.
You will be able to understand what an OverflowException is and how to trap
them at compile time.
You will know what NullPointerExceptions are and how to avoid them.
Exception Handling Chapter 5

You will be able to write your own custom exceptions that provide meaningful
information to the customer and that aid you and fellow programmers to easily
identify and resolve any issues that are raised.
You will be able to understand why you should not use exceptions to control
program flow.
You will know how to replace business rule exceptions with C# statements and
Boolean checks to control program flow.

Checked and unchecked exceptions

In unchecked mode, an arithmetic overflow is ignored. In this situation, the high-order bits
that cannot be assigned to the destination type are discarded from the result.

By default, C# operates in the unchecked context while performing non-constant

expressions at runtime. But compile-time constant expressions are always checked by
default. When an arithmetic overflow is encountered in checked mode, an
OverflowException is raised. One reason why unchecked exceptions are used is to
increase performance. Checked exceptions can decrease the performance of methods by a
small amount.

The rule of thumb is to make sure that you perform arithmetic operations in the checked
context. Any arithmetic overflow exceptions will be picked up as compile-time errors, and
you can then fix them before you release your code. That is much better than releasing your
code and then having to fix customer runtime errors.

Running code in unchecked mode is dangerous as you are making assumptions about the
code. Assumptions are not facts and they can lead to exceptions being raised at runtime.
Runtime exceptions lead to poor customer satisfaction and can produce serious follow-on
exceptions that negatively impact a customer in some way.

Allowing an application to continue running that has experienced an overflow exception is

very dangerous from a business perspective. The reason for this is that data can end up in a
non-reversible invalid state. If the data is critical customer data, then this can be
considerably costly to the business, and you don't want that on your shoulders.

Consider the following code. This code demonstrates how bad an unchecked overflow can
be in the world of customer banking:
private static void UncheckedBankAccountException()
{
var currentBalance = int.MaxValue;
Console.WriteLine($"Current Balance: {currentBalance}");

[ 110 ]
Exception Handling Chapter 5

currentBalance = unchecked(currentBalance + 1);

Console.WriteLine($"Current Balance + 1 = {currentBalance}");
Console.ReadKey();
}

Imagine the horror on this customer's face when they see that adding £1 to their bank
balance of £2,147,483,647 causes them to be in debt by -£2,147,483,648!

Now, it's time to demonstrate checked and unchecked exceptions with some code
examples. First, start a new console application and declare some variables:
static byte y, z;

The preceding code declares two bytes that we will use in our arithmetic code examples.
Now, add the CheckedAdd() method. This method will raise a checked
OverflowException if an arithmetic overflow is encountered when adding two numbers
that result in a number that is too big to be stored as a byte:
private static void CheckedAdd()
{
try
{
Console.WriteLine("### Checked Add ###");
Console.WriteLine($"x = {y} + {z}");
Console.WriteLine($"x = {checked((byte)(y + z))}");
}
catch (OverflowException oex)
{
Console.WriteLine($"CheckedAdd: {oex.Message}");
}
}

Then, write the CheckedMultiplication() method. Again, a checked

OverflowException will be raised if an arithmetic overflow is detected during the
multiplication, which results in a number that is larger than a byte:
private static void CheckedMultiplication()
{
try
{

[ 111 ]
Exception Handling Chapter 5

Console.WriteLine("### Checked Multiplication ###");

Console.WriteLine($"x = {y} x {z}");
Console.WriteLine($"x = {checked((byte)(y * z))}");
}
catch (OverflowException oex)
{
Console.WriteLine($"CheckedMultiplication: {oex.Message}");
}
}

Next, we add the UncheckedAdd() method. This method will ignore any overflow that
happens as a result of an addition, and so an OverflowException will not be raised. The
result of this overflow will be stored as a byte, but the value will be incorrect:
private static void UncheckedAdd()
{
try
{
Console.WriteLine("### Unchecked Add ###");
Console.WriteLine($"x = {y} + {z}");
Console.WriteLine($"x = {unchecked((byte)(y + z))}");
}
catch (OverflowException oex)
{
Console.WriteLine($"CheckedAdd: {oex.Message}");
}
}

And now, we add the UncheckedMultiplication() method. This method will not throw
an OverflowException when an overflow is encountered as the result of this
multiplication. The exception will simply be ignored. This will result in an incorrect
number being stored as a byte:
private static void UncheckedMultiplication()
{
try
{
Console.WriteLine("### Unchecked Multiplication ###");
Console.WriteLine($"x = {y} x {z}");
Console.WriteLine($"x = {unchecked((byte)(y * z))}");
}
catch (OverflowException oex)
{
Console.WriteLine($"CheckedMultiplication: {oex.Message}");
}
}

[ 112 ]
Exception Handling Chapter 5

Finally, it is time to modify our Main(string[] args) method so that we can initialize
the variables and execute the methods. Here, we add the maximum value for a byte to the y
variable and 2 to the z variable. Then, we run the CheckedAdd() and
CheckedMultiplication() methods, which will both generate OverflowException().
This is thrown because the y variable contains the maximum value for a byte.

So, by adding or multiplying by 2, you are exceeding the address space needed to store the
variable. Next, we will run the UncheckedAdd() and UncheckedMultiplication()
methods. Both these methods ignore overflow exceptions, assign the result to the x
variable, and disregard any bits that overflow. Finally, we print a message to the screen and
then exit when the user presses any key:
static void Main(string[] args)
{
y = byte.MaxValue;
z = 2;
CheckedAdd();
CheckedMultiplication();
UncheckedAdd();
UncheckedMultiplication();
Console.WriteLine("Press any key to exit.");
Console.ReadLine();
}

When we run the preceding code, we end up with the following output:

As you can see, when we use checked exceptions, exceptions are raised when
OverflowException is encountered. But when we use unchecked exceptions, no
exception is raised.

[ 113 ]
Exception Handling Chapter 5

It is apparent from the preceding screenshot that problems can arise from unexpected
values and that certain behaviors can arise from using unchecked exceptions. Therefore, the
rule of thumb when performing arithmetic operations must be to always use checked
exceptions.

Now, let's move on and look at a very common exception that is encountered frequently by
programmers, known as NullPointerException.

Avoiding NullPointerExceptions
NullReferenceException is a common exception that has been experienced by most
programmers. It is thrown when an attempt is made to access a property or method on a
null object.

To defend against computer program crashes, the common course of action among fellow
programmers is to use try{...}catch (NullReferenceExceptionre){...} blocks.
This is a part of defensive programming. But the problem is that, a lot of the time, the error
is simply logged and rethrown. Besides this, a lot of wasted computations are performed that
could have been avoided.

A much better way of handling ArgumentNullExceptions is to implement

ArgumentNullValidator. The parameters of a method are usually the source of a null
object. It makes sense to test the parameters of a method before they are used and, if they
are found to be invalid for any reason, to throw an appropriate Exception. In the case of
ArgumentNullValidator, you would place this validator at the top of the method and
then test each parameter. If any parameter was found to be null, then
NullReferenceException would be thrown. This would save computations and remove
the need to wrap your method's code in a try...catch block.

To make things clear, we will write ArgumentNullValidator and use it in a method to

test the method's arguments:
public class Person
{
public string Name { get; }
public Person(string name)
{
Name = name;
}
}

[ 114 ]
Exception Handling Chapter 5

In the preceding code, we have created the Person class with a single read-only property
called Name. This will be the object that we will use to pass into the example methods to
cause NullReferenceException. Next, we will create our Attribute for the validator
called ValidatedNotNullAttribibute:
[AttributeUsage(AttributeTargets.All, Inherited = false, AllowMultiple =
true)]
internal sealed class ValidatedNotNullAttribute : Attribute { }

Now that we have our Attribute, it's time to write the validator:
internal static class ArgumentNullValidator
{
public static void NotNull(string name,
[ValidatedNotNull] object value)
{
if (value == null)
{
throw new ArgumentNullException(name);
}
}
}

ArgumentNullValidator takes two arguments:

The name of the object

The object itself

The object is checked to see if it is null. If it is null, ArgumentNullException is thrown,

passing in the name of the object.

The following method is our try/catch example method. Notice that we log a message
and throw the exception. However, we don't use the declared exception parameter, and so
by rights, this should be removed. You will see this quite often in code. It is unnecessary
and should be removed to tidy the code up:
private void TryCatchExample(Person person)
{
try
{
Console.WriteLine($"Person's Name: {person.Name}");
}
catch (NullReferenceException nre)
{
Console.WriteLine("Error: The person argument cannot be null.");
throw;

[ 115 ]
Exception Handling Chapter 5

}
}

Next, we will write our example method that will use ArgumentNullValidator. We will
call it ArgumentNullValidatorExample:
private void ArgumentNullValidatorExample(Person person)
{
ArgumentNullValidator.NotNull("Person", person);
Console.WriteLine($"Person's Name: {person.Name}");
Console.ReadKey();
}

Notice that we have gone from nine lines, including braces, to only two lines. We also don't
attempt to use the value before it has been validated. All we need to do now is modify our
Main method to run the methods. Test each method by commenting out one of the methods
and running the program. When you do this, it is best to step through your code to see
what's going on.

The following is the output of running the TryCatchExample method:

The following is the output of running ArgumentNullValidatorExample:

If you study the previous screenshots carefully, you will see that we have only logged the
error once when using ArgumentNullValidatorExample. When throwing the exception
using TryCatchExample, the exception is logged twice.

The first time, we have a meaningful message, but the second time, the message is cryptic.
However, the exception that is logged by the calling method, Main, is not cryptic at all. It is,
in fact, very helpful as it shows us that the value cannot be null for the Person parameter.

[ 116 ]
Exception Handling Chapter 5

Hopefully, this section has shown you the value of checking your parameters in your
constructors and methods before you use them. By doing this, you can see how argument
validators reduce your code, thus making it more readable.

Now, we will look at implementing business rules for specific exceptions.

Business rule exceptions

Technical exceptions are exceptions that are thrown by a computer program as a result of
programmer mistakes and/or environmental issues such as there not being enough disk
space.

But business rule exceptions are different. Business rule exceptions imply that such
behavior is expected and is used to control program flow, when in fact, exceptions should
be an exception to the normal flow of the program and not the expected output of a
method.

For example, picture a person at an ATM drawing out £100 from their account that has £0
in it and does not have the ability to go overdrawn. The ATM accepts the user request to
draw £100 out, and so it issues the Withdraw(100); command. The Withdraw method
checks the balance, discovers that the account has insufficient funds, and so throws
InsufficientFundsException().

You may think that having such exceptions is a good idea as they are explicit and help
identify issues so that you can carry out a very specific action upon receiving such
exceptions – but no! This is not a good idea.

In such a scenario, when the user submits the request, the amount requested should be
checked to see if it can be withdrawn. If it can, then the transaction should go ahead, as
requested by the user. But if the validation check identifies that the transaction is unable to
go ahead, then the program should follow normal program flow to cancel the transaction
and inform the user who issued the request without raising an exception.

The withdrawal scenario we've just looked at shows that the programmer has correctly
pondered upon the normal flow of the program and the different outcomes. The program
flow has been appropriately coded using Boolean checks to allow for the successful
withdrawal transactions and to prevent disallowed withdrawal transactions.

Let's see how we would implement a withdrawal from a bank account that does not allow
an overdraft scenario using Business Rule Exceptions (BREs). Then, we'll take a look at
how we would implement the same scenario but using normal program flow instead of
employing BREs.

[ 117 ]
Exception Handling Chapter 5

Start a new console application and add two folders called

BankAccountUsingExceptions and BankAccountUsingProgramFlow. Update your
void Main(string[] args) method with the following code:

private static void Main(string[] args)

{
var usingBrExceptions = new UsingBusinessRuleExceptions();
usingBrExceptions.Run();
var usingPflow = new UsingProgramFlow();
usingPflow.Run();
}

The preceding code runs each scenario. UsingBusinessRuleExceptions() demonstrates

the use of exceptions as the expected output that's used to control program flow,
while UsingProgramFlow() demonstrates the clean way of controlling program flow
without the use of exceptional conditions.

We now need a class to hold our current account information. So, add a class called
CurrentAccount to your Visual Studio console project, as follows:

internal class CurrentAccount

{
public long CustomerId { get; }
public decimal AgreedOverdraft { get; }
public bool IsAllowedToGoOverdrawn { get; }
public decimal CurrentBalance { get; }
public decimal AvailableBalance { get; private set; }
public int AtmDailyLimit { get; }
public int AtmWithdrawalAmountToday { get; private set; }
}

The properties of this class can only be set internally or externally via the constructor. Now,
add the constructor that takes the customer identifier as the only parameter:
public CurrentAccount(long customerId)
{
CustomerId = customerId;
AgreedOverdraft = GetAgreedOverdraftLimit();
IsAllowedToGoOverdrawn = GetIsAllowedToGoOverdrawn();
CurrentBalance = GetCurrentBalance();
AvailableBalance = GetAvailableBalance();
AtmDailyLimit = GetAtmDailyLimit();
AtmWithdrawalAmountToday = 0;
}

[ 118 ]
Exception Handling Chapter 5

The current account constructor initializes all the properties. As shown in the preceding
code, some properties are initialized using methods. Let's implement each of the methods
in turn:
private static decimal GetAgreedOverdraftLimit()
{
return 0;
}

GetAgreedOverdraftLimit() returns the value of the agreed overdraft limit on the

account. In this example, it is hardcoded to zero. But in a real scenario, it would extract the
actual figure from a configuration file or other data store. This would allow non-technical
users to update the agreed overdraft limit without developers having to change the code.

GetIsAllowedToGoOverdrawn() determines if the account can be overdrawn, even if it

has not been agreed, as some banks allow. In this case, we just return false to determine
that the account is unable to go overdrawn:
private static bool GetIsAllowedToGoOverdrawn()
{
return false;
}

For the purpose of this example, we will set the user's account balance to £250 in the
GetCurrentBalance() method:

private static decimal GetCurrentBalance()

{
return 250.00M;
}

As a part of our example, we need to make sure that even if the person has £250 in their
account, but their available balance is less than that, they are unable to withdraw more than
the available balance as this would cause them to go overdrawn. To do this, we will set the
available balance to £173.64 in the GetAvailableBalance() method:
private static decimal GetAvailableBalance()
{
return 173.64M;
}

Here, in the UK, ATM machines will either allow you to withdraw a maximum of £200 or
£250. So, in the GetAtmDailyLimit() method, we will set the ATM daily limit to £250:
private static int GetAtmDailyLimit()
{

[ 119 ]
Exception Handling Chapter 5

return 250;
}

Let's write the code for our two scenarios by using business rule exceptions and normal
program flow to handle different conditions within a program.

Example 1 – handling conditions with business

rule exceptions
Add a new class to your project called UsingBusinessRuleExceptions and then add the
following Run() method:
public class UsingBusinessRuleExceptions
{
public void Run()
{
ExceedAtmDailyLimit();
ExceedAvailableBalance();
}
}

The Run() method calls two methods:

The first method is called ExceedAtmDailyLimit(). This method intentionally

exceeds the daily amount that is allowed to be withdrawn from an ATM.
ExceedAtmDailyLimit() causes ExceededAtmDailyLimitException.
Secondly, the ExceedAvailableBalance() method is called, which
intentionally causes an InsufficientFundsException. Add the
ExceedAtmDailyLimit() method:

private void ExceedAtmDailyLimit()

{
try
{
var customerAccount = new CurrentAccount(1);
customerAccount.Withdraw(300);
Console.WriteLine("Request accepted. Take cash and card.");
}
catch (ExceededAtmDailyLimitException eadlex)
{
Console.WriteLine(eadlex.Message);
}
}

[ 120 ]
Exception Handling Chapter 5

The ExceedAtmDailyLimit() method creates a new CustomerAccount method and

passes in the customer's identifier, as represented by the number 1. Then, an attempt is
made to withdraw £300. If the request is successful, then the message Request accepted.
Take cash and card. is printed to the console window. Should the request fail, then the
method traps ExceededAtmLimitException and prints the exception's message to the
console window:
private void ExceedAvailableBalance()
{
try
{
var customerAccount = new CurrentAccount(1);
customerAccount.Withdraw(180);
Console.WriteLine("Request accepted. Take cash and card.");
}
catch (InsufficientFundsException ifex)
{
Console.WriteLine(ifex.Message);
}
}

The ExceedAvailableBalance() method creates a new CurrentAccount and passes in

the customer identifier, as represented by the number 1. An attempt is then made to
withdraw £180. Since GetAvailableMethod() returns £173.64, the method causes
an InsufficientFundsException.

With that, we've seen how to manage different conditions using business rule exceptions.
Now, let's look at the proper way to manage the same conditions using normal program
flow, without the use of exceptions.

Example 2 – handling conditions with normal

program flow
Add a class called UsingProgramFlow and then add the following code to it:
public class UsingProgramFlow
{
private int _requestedAmount;
private readonly CurrentAccount _currentAccount;

public UsingProgramFlow()
{
_currentAccount = new CurrentAccount(1);

[ 121 ]
Exception Handling Chapter 5

}
}

In the constructor of the UsingProgramFlow class, we will create a new CurrentAccount

class and pass in the customer identifier. Next, we'll add the Run() method:
public void Run()
{
_requestedAmount = 300;
Console.WriteLine($"Request: Withdraw {_requestedAmount}");
WithdrawMoney();
_requestedAmount = 180;
Console.WriteLine($"Request: Withdraw {_requestedAmount}");
WithdrawMoney();
_requestedAmount = 20;
Console.WriteLine($"Request: Withdraw {_requestedAmount}");
WithdrawMoney();
}

The Run() method sets the _requestedAmount variable three times. Each time it does this,
a message is printed stating the withdrawn amount on the console window before calling
the WithdrawMoney() method. Now, add the ExceedsDailyLimit() method:
private bool ExceedsDailyLimit()
{
return (_requestedAmount > _currentAccount.AtmDailyLimit)
|| (_requestedAmount + _currentAccount.AtmWithdrawalAmountToday >
_currentAccount.AtmDailyLimit);
}

The ExceedDailyLimit() method returns true if _requestedAmount exceeds the daily

ATM withdrawal limit. Otherwise, it returns false. Now, add the
ExceedsAvailableBalance() method:

private bool ExceedsAvailableBalance()

{
return _requestedAmount > _currentAccount.AvailableBalance;
}

The ExceedsAvailableBalance() method returns true if the requested amount is more

than is available for withdrawal. Finally, we come to the last method, called
WithdrawMoney():

private void WithdrawMoney()

{
if (ExceedsDailyLimit())
Console.WriteLine("Cannot exceed ATM Daily Limit. Request

[ 122 ]
Exception Handling Chapter 5

denied.");
else if (ExceedsAvailableBalance())
Console.WriteLine("Cannot exceed available balance. You have no
agreed
overdraft facility. Request denied.");
else
Console.WriteLine("Request granted. Take card and cash.");
}

The WithdrawMoney() method does not use BREs to control the program flow. Instead,
this method calls Boolean validation methods that determine the program flow. If
_requestedAmount exceeds the ATM daily limit, as determined by the call to
ExceedsDailyLimit(), then the request is denied. Otherwise, the next check is carried
out to see if _requestedAmount is more than AvailableBalance. If it is, then the request
is rejected. If not, then the code is executed that grants the request.

I hope you can see that it makes more sense to control the flow of a program using the
available logic rather than expecting exceptions to be thrown. The code is a lot cleaner and
more correct. Exceptions should be reserved for exceptional circumstances that are not a
part of the business requirements.

When proper exceptions are raised in the correct manner, it is important for them to be
meaningful. Cryptic error messages are no good for anyone and can actually add
unnecessary stress for end users or developers. Now, we are going to look at providing
meaningful information in any of the exceptions that are raised by our computer programs.

Exceptions should provide meaningful

information
Critical errors that state "There is no error" and then kill a program are just not useful at all.
I have experienced the actual "There is no error" critical exception first hand. It is a critical
exception that stops an application from working. Yet the message is informing us that
there is no error. Well, if there is no error, then why has a critical exception warning
appeared on the screen? And why am I unable to continue using the application?
Obviously, for the critical exception to be raised, there must be a critical exception
somewhere that occurred. But where and why?

[ 123 ]
Exception Handling Chapter 5

What makes such exceptions even more annoying is when they are deep-rooted in the
framework or library that you are using (which you have no control over), and where you
have no access to the source code. Such exceptions have caused programmers to say
negative things out of frustration. I've been guilty of this and I've experienced fellow
colleagues do the same. One of the main reasons for the frustration is the unhelpful fact that
the code has raised an error and the user or programmer has been informed, but there is no
helpful information to suggest what the problem is or where to look or even what remedial
action to take.

Exceptions must provide information that is human-friendly, especially to the technically

challenged. During my time developing dyslexia testing and assessment software, I have
worked with many teachers and IT technicians.

It can be said that many IT technicians and teachers at all levels of ability have often been
clueless when it comes to responding to software exception messages.

One error that has perplexed many of the end users of the software I've supported has been
Error 76: Path not found. This is an old Microsoft exception that has been around as far
back as Windows 95, and that still exists today. For the end user of the software that raises
this exception, the error message is totally useless. It would be useful for the end user to
know what file and location cannot be found and to know what steps to take to remedy the
situation.

A potential solution would be to implement the following steps:

1. Check for the existence of the location.

2. If the location does not exist or access is denied, then display the file save or open
dialog as needed.
3. Save the user-selected location to a configuration file for future use.
4. On subsequent runs of the same code, use the location set by the user.

But if you were to stay with the error message, then you should at least provide the name
of the location and/or file that is missing.

With that said, it is now time to look at how we can build our own exceptions to provide
just the right amount of information that will be useful to the end user and to the
programmer. But take note: you must be careful not to disclose sensitive information or
data.

[ 124 ]
Exception Handling Chapter 5

Building your own custom exceptions

Microsoft .NET Framework already has a good number of exceptions that can be raised that
you are able to trap. But there may be instances where you'll require a custom exception
that provides more detailed information or that is more end user friendly in its
terminology.

So, we are now going to look at what the requirements are for building our own custom
exceptions. It is surprisingly simple to build your own custom exception. All you have to
do is give your class a name that ends with Exception and inherit from
System.Exception. Then, you need to add three constructors, as shown in the following
code example:
public class TickerListNotFoundException : Exception
{
public TickerListNotFoundException() : base()
{
}

public TickerListNotFoundException(string message)

: base(message)
{
}

public TickerListNotFoundException(
string message,
Exception innerException
)
: base(message, innerException)
{
}
}

TickerListNotFoundException inherits from the System.Exception class. It contains

three mandatory constructors:

A default constructor
A constructor that accepts a string of text for the exception message
A constructor that accepts a string of text for the exception message and an
Exception object for the inner exception

[ 125 ]
Exception Handling Chapter 5

We are now going to write and execute three methods that will use each of our custom
exception's constructors. You will be able to clearly see the benefit of using custom
exceptions to create more meaningful exceptions:
static void Main(string[] args)
{
ThrowCustomExceptionA();
ThrowCustomExceptionB();
ThrowCustomExceptionC();
}

The preceding code shows our updated Main(string[] args) method, which has been
updated to execute our three methods. These will test each of our custom exception's
constructors in turn:
private static void ThrowCustomExceptionA()
{
try
{
Console.WriteLine("throw new TickerListNotFoundException();");
throw new TickerListNotFoundException();
}
catch (Exception tlnfex)
{
Console.WriteLine(tlnfex.Message);
}
}

The ThrowCustomExceptionA() method throws a new

TickerListNotFoundException by using the default constructor. When you run the
code, the message that's printed to the console window informs the user that a
CH05_CustomExceptions.TickerListNotFoundException has been thrown:

private static void ThrowCustomExceptionB()

{
try
{
Console.WriteLine("throw new
TickerListNotFoundException(Message);");
throw new TickerListNotFoundException("Ticker list not found.");
}
catch (Exception tlnfex)
{
Console.WriteLine(tlnfex.Message);
}
}

[ 126 ]
Exception Handling Chapter 5

ThrowCustomExceptionB() throws a new TickerListNotFoundException by using

the constructor that accepts a text message. In this case, the end user is informed that the
ticker list hasn't been found:
private static void ThrowCustomExceptionC()
{
try
{
Console.WriteLine("throw new TickerListNotFoundException(Message,
InnerException);");
throw new TickerListNotFoundException(
"Ticker list not found for this exchange.",
new FileNotFoundException(
"Ticker list file not found.",
@"F:\TickerFiles\LSE\AimTickerList.json"
)
);
}
catch (Exception tlnfex)
{
Console.WriteLine($"{tlnfex.Message}\n{tlnfex.InnerException}");
}
}

Finally, the ThrowCustomExceptionC() method throws a

TickerListNotFoundException by using the constructor that takes a text message and
inner exception. In our example, we provide a meaningful message stating that the ticker
list has not been found for this exchange. The inner FileNotFoundException expands
upon this by providing the name of the specific file that was not found, which happens to
be the ticker list of Aim companies on the London Stock Exchange (LSE).

Here, we can see that there are genuine advantages to creating your own custom
exceptions. But in most cases, using the intrinsic exceptions within .NET Framework should
suffice. The main benefit of custom exceptions is that they're more meaningful exceptions
that aid with debugging and resolution.

Here is a brief list of C# exception handling best practices:

Use try/catch/finally blocks to recover from errors or release resources.

Handle common conditions without throwing exceptions.
Design classes so that exceptions can be avoided.
Throw exceptions instead of returning an error code.
Use the predefined .NET exception types.
End exception class names with the word Exception.

[ 127 ]
Exception Handling Chapter 5

Include three constructors in custom exception classes.

Ensure that exception data is available when code executes remotely.
Use grammatically correct error messages.
Include a localized string message in every exception.
In custom exceptions, provide additional properties as needed.
Place throw statements so that the stack trace will be helpful.
Use exception builder methods.
Restore state when methods don't complete due to exceptions.

Now, it's time to summarize what we have learned in regard to exception handling.

Summary
In this chapter, you learned about checked exceptions and unchecked exceptions. Checked
exceptions prevent arithmetic overflow conditions from entering any production code as
they are trapped at compile time. Unchecked exceptions go unchecked at compile time and
can often make it into production code. This can lead to some hard-to-track-down bugs in
your code through unexpected data values and even result in exceptions being thrown that
cause your programs to crash.

You then learned about the common NullPointerException and how to validate
parameters that have been passed in using custom Attribute and Validator classes,
which are placed at the top of your methods. These allow you to provide meaningful
feedback when validation fails. This leads to more robust programs in the long run.

Then, we discussed using BREs to control program flow. You were shown how to control
the program flow by expecting exceptional output. Then, you saw how to achieve better
control over the flow of computer code without using exceptions by using conditional
checks.

The discussion then moved onto the importance of providing meaningful exception
messages and how this can be achieved; that is, by writing your own custom exceptions
that inherit from the Exception class and implement the required three parameters.
Through the examples provided, you learned how to use your custom exceptions and how
they aid better debugging and resolution.

So, now, it is time to put what you've learned to the test by answering some questions.
There is also further reading for you to do if you wish to expand upon what you have
learned in this chapter.

[ 128 ]
Exception Handling Chapter 5

In the next chapter, we will be looking at unit testing and how to write your tests first so
that they fail. Then, we will write just enough code for the tests to pass and refactor the
working code before moving on to the next unit test.

Questions
1. What is a checked exception?
2. What is an unchecked exception?
3. What is an arithmetic overflow exception?
4. What is a NullPointerException?
5. How can you validate null parameters to improve your overall code?
6. What does BRE stand for?
7. Are BREs good or bad practice, and why do you think that?
8. What is the alternative to BREs, is it good or bad, and why do you think that?
9. How can you provide meaningful exception messages?
10. What are the requirements for writing your own custom exceptions?

Further reading
https://docs.microsoft.com/en-us/dotnet/standard/exceptions/: This is the
official documentation for handling and throwing exceptions in .NET.
https://reflectoring.io/business-exceptions/: The author of this article
provides five reasons why BREs are a bad idea after originally believing they
were a good idea. There is extra information in this article that was not covered
in this chapter.
https://docs.microsoft.com/en-us/dotnet/standard/exceptions/best-
practices-for-exceptions: The best practices from Microsoft in regard to C#
exception handling, along with code examples and explanations.

[ 129 ]
6
Unit Testing
Previously, we looked at exception handling, how to implement it properly, and how this
can be useful to the customer and the programmer when issues occur. In this chapter, we
will look at how programmers can implement their own quality assurance (QA) to provide
quality code that is robust and less likely to generate exceptions in production.

We start by looking at why we should test our own code, and what makes a good test. We
then look at several testing tools that are available to C# programmers. Then, we move on
to the three pillars of unit testing that are Fail, Pass, and Refactor. Finally, we look at
redundant unit tests and why they should be removed.

In this chapter, we will cover the following topics:

Understanding the reasons for a good test

Understanding the testing tools
TDD methodology practice – fail, pass, and refactor
Removing redundant tests, comments, and dead code

By the end of this chapter, you will have gained the following skills:

Be able to describe the benefits of good code

Be able to describe potential negatives that can arise from not unit testing
Be able to install and use MSTest to write and run unit tests
Be able to install and use NUnit to write and run unit tests
Be able to install and use Moq to write fake (mock) objects
Be able to install and use SpecFlow to write software that adheres to customer
specifications
Be able to write tests that fail, then make them pass, and then perform any
necessary refactoring
Unit Testing Chapter 6

Technical Requirements
To access the code files of this chapter, you can visit this link: https://github.com/
PacktPublishing/Clean-Code-in-C-/tree/master/CH06.

Understanding the reasons for a good test

As a programmer, it is nice to work on a new development project that you find interesting,
especially if you are highly motivated to do so. But it can be extremely frustrating if you get
called away to work on a bug instead. It can be worse if it is not your code, and you don't
have the full understanding behind the code. It is even worse still if it is your own code and
you have that "What was I thinking?" moment! The more you get called away from new
development to perform maintenance on existing code, the more you begin to appreciate
the need for unit testing. As this appreciation grows, you begin to see the real benefits of
learning testing methodologies and techniques such as Test-Driven Development (TDD)
and Behavioral-Driven Development (BDD).

When you've spent a period of time working as a maintenance programmer on other

people's code, you get to see the good, the bad, and the ugly. Such code can be a positive
education that opens your eyes to a better way of programming by understanding what not
to do and why not to do it. The bad code can make you shout No. Just no! and the ugly code
can cause your eyes to bleed and your mind to go numb.

Dealing directly with customers, providing them with technical support, you see just how
crucial a good customer experience is to the success of the business. Conversely, you also
get to see how a bad customer experience can lead to some very frustrated, angry, and
extremely foul-mouthed customers; and how quickly sales can be lost due to customer
refunds and loss of customers because of very harmful customer rants on social media and
review sites.

As a tech lead, it is your responsibility to perform technical code reviews to ensure that staff
adhere to the company's coding guidelines and policies, triage bugs, and assist the project
manager in managing the people you are responsible for leading. It is important as a tech
lead to be good at high-level project management, requirements gathering and analysis,
architectural design, and clean programming. You also need to have good people skills.

[ 131 ]
Unit Testing Chapter 6

Your project manager is only interested in delivering a project on time and to budget
according to the needs of the business. They really don't care about how you code the
software, only that you get it done on time and to the agreed budget. Most importantly,
they care that the released software exactly matches what the business asks for – no more
and no less – and that the software is to a very high and professional standard, as the
quality of the code can equally boost or destroy a company brand. When a project manager
is harsh with you, you know the business is putting them under increased pressure. And so
that pressure trickles down to you.

As a tech lead, you are sandwiched between the project manager and the team working on
the project. In your everyday work, you will be running scrum meetings and dealing with
problems. Those problems may be the coders needing resources from the analysts, testers
waiting for bugs to be fixed by the developers, and so on. But the most difficult job will be
to perform peer code reviews and provide constructive feedback that gets the desired
results without offending people. That is why you should take clean coding very seriously,
because if you criticize a person's code, you open yourself up for a backlash if your own
code is not up to scratch. But also, you will be the one to get it in the neck from the project
manager if the software fails testing or goes out with loads of bugs.

That is why, as a tech lead, it is a good idea for you to encourage TDD. The best way to do
that is by leading by example. Now I know that even degree-educated and experienced
programmers can be very stand-offish to TDD. One of the most common reasons is that it
can be hard to learn and put into practice, and appear to be more time consuming,
especially when code becomes more complex. I have experienced these kinds of objections
from my colleagues who prefer not to unit test.

But as a programmer, if you want to be truly confident (such that once you've written a
piece of code, you can be confident in its quality and that it will not be returned to you to
fix your own bugs), then TDD is a fantastic way to up your game as a programmer. When
you learn to test first before you start programming, it soon becomes habitual. Such a habit,
as a programmer, is very useful and beneficial to you, especially when there comes a time
to find a new position, as many employment opportunities advertise for people with TDD
or BDD experience.

[ 132 ]
Unit Testing Chapter 6

Another thing to consider while writing code is that bugs in a simple, non-critical note-
taking app are not the end of the world. But what if you work in the defense or health
sectors? Consider a weapon of mass destruction that has been programmed to go in a
specific direction to hit a specific target in enemy territory, but something goes wrong, and
the missile aims for civilian populations that belong to your allies. Or, consider what would
happen if you had a loved one that was on critical life support that died because of a bug in
the software of the medical equipment that was your own fault. Then, what about some
safety software going wrong on a passenger jet flying over a populated area that causes the
plane to crash into the ground, killing people on the plane and on the ground?

The more critical the software, the more the use of unit testing techniques (such as TDD and
BDD) needs to be taken seriously. We will be discussing BDD and TDD tools later in this
chapter. When writing software, think about how you would be affected if you were the
customer and something went wrong with the code you are writing. How would it affect
your family, friends, and colleagues? Also, think of the moral and legal implications if you
were responsible for a critical failure.

It is important to understand why, as a programmer, you should learn to test your own
code. It is true what they say that "programmers should never test their own code". But it is only
true in the context where the code is finished and ready for testing before it goes into
production. So while the code is still being programmed, programmers should always be
testing their own code. Yet some businesses are so time-constrained that proper QA is often
sacrificed so that the business can be the first to market.

It may be very important for a business to be the first to market, but first impressions count.
If a business is first to market, and the product has some serious flaws that become globally
broadcast, this can have a long-lasting negative impact on a business. So you must think
very carefully as a programmer and do your best to ensure that if the software has flaws,
you are not the one responsible. When things go wrong in a business, heads will roll. And
in Teflon Management, the managers will pass the guilt for driving ridiculous
deadlines from themselves all the way down to the programmers that had to meet the
deadline and make sacrifices to do so.

So you see, it is very important as a programmer that you test your code and test it often,
especially prior to releasing it to the testing team. That is why you are actively encouraged
to transition into the mindset and habitual behavior of writing your tests first, based upon
the specification that you currently implementing. Your tests should fail to start with. You
then write only enough code to get the tests to pass, and then you refactor your code as you
need to.

[ 133 ]
Unit Testing Chapter 6

It is hard to get started with TDD or BDD. But once you get the hang of it, TDD and BDD
become second nature. And you will probably find that in the long term, you are left with
cleaner code that is easy to read and maintain. You may also find that your confidence in
your ability to modify the code without breaking it may also be greatly improved.
Obviously, there is more code in the sense that you have the production method and the
test method(s). But you may actually end up writing less code overall, as you will not be
adding extra code that you think may be needed!

Picture yourself at your computer with a software specification that you have to translate
into working software. A bad habit that many programmers have, and that I've been guilty
of in the past, is that they jump straight into coding without doing any real design work. In
my experience, this actually prolongs the time it takes to develop a piece of code and can
often lead to more bugs and code that is hard to maintain and extend. In fact, although it
appears to be counter-intuitive to some programmers, proper planning and design actually
speed up coding, especially when you factor in maintenance and extensions.

This is where the test team comes in. Before we go any further, let's describe use cases, test
designs, test cases, and test suites, and how they relate to one another.

A use case explains the process flow for a single operation, such as adding a customer
record. A test design will comprise one or more test cases that test for different scenarios
that could take place for the single use case. The test cases may be carried out manually, or
they may be automated tests that are executed by a test suite. A test suite is a piece of
software used to discover and run tests and to report their outcomes to an end user. The
writing of use cases will be the role of the business analyst. As for the test design, test cases,
and test suite, these will be the responsibility of the dedicated test team. Developers need
not be concerned with putting together the use cases, test designs of test cases, and their
execution in the test suite. Developers must focus on writing and using their unit tests to
write code that fails, then runs, and is then refactored as necessary.

Software testers collaborate with programmers. This collaboration normally starts at project
inception, and continues right through to the end. Both the development team and testing
team will collaborate by sharing test cases for each product backlog item. This process
normally consists of writing test cases. For the tests to pass, they will have to meet test
criteria. These test cases will normally be run using a combination of manual testing and
some test suite automation.

[ 134 ]
Unit Testing Chapter 6

During the development phases, the testers write their QA tests and the developers write
their unit tests. When developers submit their code to the test team, the test team will run
through their battery of tests. The outcome of those tests will be fed back to the developers
and the project stakeholders. If problems are encountered, this is known as technical debt.
The development team will have to factor in time to address the issues raised by the test
team. When the test team confirms that the software has been completed to the required
level of quality, then the code is passed on to infrastructure to release into production.

Assuming we are starting a brand new project (also known as a greenfield project), we
would select the appropriate project type and tick the option to include a test project. This
would create a solution that consists of our main project and the test project.

The type of project that we create and any features of projects to be implemented will be
dependent upon use cases. Use cases are used during system analysis to identify, confirm,
and organize software requirements. From use cases, test cases can be assigned to the
acceptance criteria. As a programmer, you can take these use cases and their test cases to
build up your own unit tests for each test case. Your tests are then run as part of a test suite.
In Visual Studio 2019, you can access the Test Explorer from the View | Test Explorer
menu. When you build your project, tests will be discovered. When tests are discovered,
they are viewed in the Test Explorer. You can then run and/or debug your tests in the Test
Explorer.

It is worth noting at this stage that it will be the responsibility of the testers and not the
developers to design tests and come up with a suitable number of test cases. They are also
responsible for QA once the software leaves the hands of the developers. But it is still the
responsibility of the developer to unit test their code, and this is where test cases can be a
real help and motivation for writing unit tests in your code.

When the solution is created, the first thing you do is open the provided test class. In that
test class, you write the pseudocode for what you must accomplish. You then go step by
step through the pseudocode and add your test methods that test each step that must be
accomplished in order to reach your goal of a completed software project. Each test method
that you write is written to fail. You then write just enough code to pass the test. Then, once
the test passes, you refactor your code before progressing to the next test. So, you can see
that unit testing is not rocket science. But what does it take to write a good unit test?

Any code that is under test will be expected to provide a specific function. A function takes
in input and produces output.

[ 135 ]
Unit Testing Chapter 6

In a normally functioning computer program, a method (or function) will have an acceptable
range of inputs and outputs, and an unacceptable range of inputs and outputs. And so the
perfect unit test will test the lowest acceptable value, the highest acceptable value, and will
provide test cases that are outside of the acceptable range of values both high and low.

Unit tests must be atomic, which means that they should only test one thing. Since methods
can be chained together in the same class and even across multiple classes in multiple
assemblies, it is often useful to provide fake or mock objects for the classes under test to
keep them atomic. The output must determine whether it passes or fails. Good unit tests
must never be inconclusive.

The result of a test should be repeatable, in that it either always passes or always fails in
given conditions. That is, the same test run over and over again should not have different
outcomes each time it is run. If it does, then it is not repeatable. Unit tests should not have
to rely on other tests being run before them, and they should be isolated from other
methods and classes. You should also aim for unit tests that run in milliseconds. Any test
that takes one second or more to run is taking too long. If code takes longer than a second,
then you should consider refactoring or implementing a mock object for testing. And since
we are busy programmers, unit tests should be easy to set up and not require a lot of
coding or configuration. The following diagram shows the unit testing life cycle:

[ 136 ]
Unit Testing Chapter 6

We'll be writing unit tests and mock objects during this chapter. But before we do, we'll
need to look at some of the tools that are available to us as C# programmers.

Understanding the testing tools

The testing tools we'll be looking at within Visual Studio are MSTest, NUnit, Moq, and
SpecFlow. Each testing tool creates a console application and the relevant test project.
NUnit and MSTest are unit testing frameworks. NUnit is much older than MSTest, and so
has a more mature and full-featured API compared to MSTest. I personally prefer NUnit
over MSTest.

Moq is different from MSTest and NUnit as it is not a testing framework but a mocking
framework. A mocking framework replaces the real classes in your project with mock (fake)
implementations that are used for testing purposes. You can use Moq together with MSTest
or NUnit. And finally, SpecFlow is a BDD framework. You start by writing a feature in a
feature file using business language that the user and the techy alike will understand. Then
a step file is generated for that feature. The step file contains the methods as steps necessary
to implement that feature.

By the end of this chapter, you will understand what each tool does and will be able to use
them in your own projects. So, let's get started by looking at MSTest.

MSTest
In this section, we will install and configure the MSTest Framework. We will write a test
class with test methods and initialize it. We will perform assembly setup and cleanup, class
cleanup, and method cleanup, and perform assertions.

To install the MSTest Framework from the command line in Visual Studio, you will need to
open the Package Manager Console via Tools | NuGet Package Manager | Package
Manager Console:

[ 137 ]
Unit Testing Chapter 6

Then, run the following three commands to install the MSTest Framework:
install-package mstest.testframework
install-package mstest.testadapter
install-package microsoft.net.tests.sdk

[ 138 ]
Unit Testing Chapter 6

Alternatively, you can add a new project and select Unit Test Project (.NET Framework)
from the Context | Add menu in the Solution Explorer. See the screenshot that
follows. When naming test projects, the accepted standard is in the form
of <ProjectName>.Tests. This helps to associate them with the tests and distinguish
them from the project that is under test:

[ 139 ]
Unit Testing Chapter 6

The following code is the default unit test code that is generated when you add an MSTest
project to your solution. As you can see, the class imports
the Microsoft.VisualStudio.TestTools.UnitTesting namespace.
The [TestClass] attribute identifies to the MS Test Framework that this class is a test
class. The [TestMethod] attribute marks the method as a test method. All classes that have
the [TestMethod] attribute will appear in the test player.
The [TestClass] and [TestMethod] attributes are mandatory:
using Microsoft.VisualStudio.TestTools.UnitTesting;

namespace CH05_MSTestUnitTesting.Tests
{
[TestClass]
public class UnitTest1
{
[TestMethod]
public void TestMethod1()
{
}
}
}

There are other methods and attributes that can optionally be combined to produce a
complete test execution workflow. These include [AssemblyInitialize],
[AssemblyCleanup], [ClassInitialize], [ClassCleanup], [TestInitialize], and
[TestCleanup]. As their names imply, the initialization attributes are used to perform any
initialization at the assembly, class, and method level prior to tests being run. Likewise, the
cleanup attributes run at the method, class, and assembly level after tests have been run to
perform any necessary cleanup operations. We will look at each in turn and add them to
your project as we will see its order of execution when we run the final code.

The WriteSeparatorLine() method is a helper method for the purpose of separating our
testing method outputs. This will help us to more easily follow what's going on with our
test class:
private static void WriteSeparatorLine()
{
Debug.WriteLine("--------------------------------------------------");
}

Optionally, assign the [AssemblyInitialize] attribute to execute code before the tests
are executed:
[AssemblyInitialize]
public static void AssemblyInit(TestContext context)

[ 140 ]
Unit Testing Chapter 6

{
WriteSeparatorLine();
Debug.WriteLine("Optional: AssemblyInitialize");
Debug.WriteLine("Executes once before the test run.");
}

Then, you can optionally assign the [ClassInitialize] attribute to execute code once
before the tests are executed:
[ClassInitialize]
public static void TestFixtureSetup(TestContext context)
{
WriteSeparatorLine();
Console.WriteLine("Optional: ClassInitialize");
Console.WriteLine("Executes once for the test class.");
}

Then, run the setup code before each unit test by assigning
the [TestInitialize] attribute to a setup method:
[TestInitialize]
public void Setup()
{
WriteSeparatorLine();
Debug.WriteLine("Optional: TestInitialize");
Debug.WriteLine("Runs before each test.");
}

When you have finished your test run, you can optionally assign
the [AssemblyCleanup] attribute to perform any necessary cleanup operations:
[AssemblyCleanup]
public static void AssemblyCleanup()
{
WriteSeparatorLine();
Debug.WriteLine("Optional: AssemblyCleanup");
Debug.WriteLine("Executes once after the test run.");
}

The optional method marked as [ClassCleanup] runs once after all tests in the class have
been executed. You cannot guarantee when this method will run, as it may not run
immediately after the execution of all tests:
[ClassCleanup]
public static void TestFixtureTearDown()
{
WriteSeparatorLine();
Debug.WriteLine("Optional: ClassCleanup");

[ 141 ]
Unit Testing Chapter 6

Debug.WriteLine("Runs once after all tests in the class have been

executed.");
Debug.WriteLine("Not guaranteed that it executes instantly after all
tests the class have executed.");
}

To perform clean up operations after each test has been run, apply
the [TestCleanup] attribute to the test cleanup method:
[TestCleanup]
public void TearDown()
{
WriteSeparatorLine();
Debug.WriteLine("Optional: TestCleanup");
Debug.WriteLine("Runs after each test.");
Assert.Fail();
}

Now that our code is in place, build it. Then, from the Test menu, select Test Explorer. You
should see the following test in the Test Explorer. As you can from the following
screenshot, the test has not yet been run:

So, let's run our only test. Oh no! Our test has failed, as shown in the following screenshot:

[ 142 ]
Unit Testing Chapter 6

Update the TestMethod1() code as shown in the following snippet, and then run the test
again:
[TestMethod]
public void TestMethod1()
{
WriteSeparatorLine();
Debug.WriteLine("Required: TestMethod");
Debug.WriteLine("A test method to be run by the test runner.");
Debug.WriteLine("This method will appear in the test list.");
Assert.IsTrue(true);
}

You see that the test has passed in the Test Explorer, as shown in the screenshot that
follows:

So, from the previous screenshots, you can see that tests that have not been executed are
blue, tests that fail are red, and tests that pass are green. From Tools | Options | Debugging
| General, select Redirect all Output Window text to the Immediate Window. Then, select
Run | Debug All Tests.

As you run through the tests and the output is printed to Immediate Window, it will
become apparent in what order the attributes are being executed. The following screenshot
shows the output from our test methods:

[ 143 ]
Unit Testing Chapter 6

As you have seen already, we have used two Assert methods—these being
Assert.Fail() and Assert.IsTrue(true). The Assert class is very useful and so it
pays to be aware of the methods available in the class for unit testing. These available
methods are listed and described as follows:

Methods Description
Tests whether the specified values are equal and throws an
Assert.AreEqual()
exception if the two values are not equal.
Tests whether the specified values are unequal and throws
Assert.AreNotEqual()
an exception if the two values are equal.
Tests whether the specified objects refer to different objects
Assert.ArtNotSame() and throws an exception if the two inputs refer to the same
object.
Tests whether the specified objects both refer to the same
Assert.AreSame() object and throws an exception if the two inputs do not refer
to the same object.
This object will always throw with Assert.Fail. Hence,
Assert.Equals()
we can use Assert.AreEqual instead.
Assert.Fail() Throws an AssertFailedException exception.
Throws an
Assert.Inconclusive()
AssertInconclusiveException exception.

[ 144 ]
Unit Testing Chapter 6

Tests whether the specified condition is false and throws an

Assert.IsFalse()
exception if the condition is true.
Tests whether the specified object is an instance of the
Assert.IsInstanceOfType() expected type and throws an exception if the expected type
is not in the inheritance hierarchy of the object.
Tests whether the specified object is an instance of the
Assert.IsNotInstanceOfType() wrong type and throws an exception if the specified type is
in the inheritance hierarchy of the object.
Tests whether the specified object is non-null and throws an
Assert.IsNotNull()
exception if it is null.
Tests whether the specified object is null and throws an
Assert.IsNull()
exception if it is not null.
Tests whether the specified condition is true and throws an
Assert.IsTrue()
exception if the condition is false.
Determines whether the specified object instances are the
Assert.ReferenceEquals()
same instance.
Assert.ReplaceNullChars() Replaces null characters ('\0') with "\\0".
Assert.That() Gets the singleton instance of the Assert functionality.
Tests whether the code specified by delegate action throws
given an exception of type T (and not a derived type) and
throws AssertFailedException if the code does not
Assert.ThrowsException() throw an exception, or throws an exception of a type other
than T. In simple words, this takes a delegate and asserts
that it throws the expected exception with the expected
message.
Tests whether the code specified by delegate action throws
given the exception of type T (and not a derived type) and
Assert.ThrowsExceptionAsync() throws AssertFailedException if the code does not
throw an exception, or throws an exception of a type other
than T.

Now that we have had a look at MSTest, it is time to look at NUnit.

NUnit
If NUnit is not installed for Visual Studio, then download and install it via Extensions |
Manage Extensions. After that, create a new NUnit Test Project (.NET Core). The following
code contains the default class created by NUnit, called Tests:
public class Tests
{
[SetUp]
public void Setup()

[ 145 ]
Unit Testing Chapter 6

{
}

[Test]
public void Test1()
{
Assert.Pass();
}
}

As you can see from the Test1 method, the test methods also use an Assert class, as does
MSTest for testing assertions in code. The NUnit Assert class makes the following methods
available to us (note that methods marked as [NUnit] in the following table are specific to
NUnit; all others are also present in MSTest):

Methods Description
Verifies that two items are equal. If they are not equal, then an
Assert.AreEqual()
exception is thrown.
Verifies that two items are not equal. If they are equal, then an
Assert.AreNotEqual()
exception is thrown.
Verifies that two objects do not refer to the same object. If they
Assert.AreNotSame()
do, then an exception is thrown.
Verifies that two objects refer to the same object. If they don't,
Assert.AreSame()
then an exception is thrown.
[NUnit] Applies a constraint to an actual value, succeeding if
the constraint is satisfied and throwing an assertion exception
Assert.ByVal()
on failure. Used as a synonym for That in rare cases where a
private setter causes a Visual Basic compilation error.
[NUnit] Verifies that a delegate throws an exception when
Assert.Catch()
called and returns it.
Assert.Contains() [NUnit] Verifies whether a value is contained in a collection.
Assert.DoesNotThrow() [NUnit] Verifies that a method does not throw an exception.
Assert.Equal() [NUnit] Do not use. Use Assert.AreEqual() instead.
Assert.Fail() Throws an AssertionException.
[NUnit] Verifies a condition is false. Throws an exception if the
Assert.False()
condition is true.
[NUnit] Verifies that the first value is greater than the second
Assert.Greater()
value. Throws an exception if it is not.
[NUnit] Verifies that the first value is greater than or equal to
Assert.GreaterOrEqual()
the second value. Throws an exception if it is not.
[NUnit] Throws IgnoreException with the message and
Assert.Ignore() arguments that are passed in. This causes the test to be
reported as ignored.

[ 146 ]
Unit Testing Chapter 6

Throws InconclusiveException with the message and

Assert.Inconclusive() arguments that are passed in. This causes the test to be
reported as inconclusive.
[NUnit] Verifies that an object may be assigned a value of a
Assert.IsAssignableFrom()
given type.
[NUnit] Verifies whether a value such as a string or collection
Assert.IsEmpty()
is empty.
Verifies whether a condition is false. Throws an exception if it
Assert.IsFalse()
is true.
Assert.IsInstanceOf() [NUnit] Verifies that an object is an instance of a given type.
[NUnit] Verifies that the value is not a number. If it is, then an
Assert.NAN()
exception is thrown.

Assert.IsNotAssignableFrom() [NUnit] Verifies that an object is not assignable from a given

type.
Assert.IsNotEmpty() [NUnit] Verifies that a string or collection is not empty.
[NUnit] Verifies that the object is not an instance of a given
Asserts.IsNotInstanceOf()
type.
Verifies that an object is not null. If it is, then an exception is
Assert.InNotNull()
thrown.
Verifies that an object is null. If it is not, then an exception is
Assert.IsNull()
thrown.
Verifies that a condition is true. If it is false, then an exception
Assert.IsTrue()
is thrown.
[NUnit] Verifies that the first value is less than the second
Assert.Less()
value. If not, then an exception is thrown.
[NUnit] Verifies that the first value is less than or equal to the
Assert.LessOrEqual()
second value. If not, then an exception is thrown.
[NUnit] Wraps code containing a series of assertions, which
Assert.Multiple() should all be executed, even if they fail. Failed results are saved
and reported at the end of the code block.
[NUnit] Verifies that a number is negative. If not, then an
Assert.Negative()
exception is thrown.
[NUnit] Verifies that an object is not null. If it is null, then an
Assert.NotNull()
exception is thrown.
[NUnit] Verifies that a number is not zero. If it is zero, then an
Assert.NotZero()
exception is thrown.
[NUnit] Verifies that an object is null. If not, then an exception
Assert.Null()
is thrown.
[NUnit] Throws SuccessException with the message and
Assert.Pass() arguments that are passed in. This allows a test to be cut short,
with a result of success returned to NUnit.
Assert.Positive() [NUnit] Verifies that a number is positive.

[ 147 ]
Unit Testing Chapter 6

[NUnit] Do not use. Throws

Assert.ReferenceEquals()
InvalidOperationException.
Verifies that a condition is true. If not, then an exception is
Assert.That()
thrown.
Verifies that a delegate throws a particular exception when it is
Assert.Throws()
called.
[NUnit] Verifies that a condition is true. If not, then an
Assert.True()
exception is called.
[NUnit] Issues a warning using the message and arguments
Assert.Warn()
provided.
Assert.Zero() [NUnit] Verifies that a number is zero.

The NUnit life cycle begins with the TestFixtureSetup that is executed once before the
first test SetUp. Then, SetUp is executed before each test. After each test has executed,
TearDown is executed. And finally, TestFixtureTearDown is executed once after the last
test TearDown. We are now going to update the Tests class so that we can debug and see
the NUnit life cycle in action:
using System;
using System.Diagnostics;
using NUnit.Framework;

namespace CH06_NUnitUnitTesting.Tests
{
[TestFixture]
public class Tests : IDisposable
{
public TestClass()
{
WriteSeparatorLine();
Debug.WriteLine("Constructor");
}

public void Dispose()

{
WriteSeparatorLine();
Debug.WriteLine("Dispose");
}
}
}

We have added the [TestFixture] to the class and implemented the IDisposable
interface. The [TextFixture] attribute is optional for non-parameterized and non-generic
fixtures. A class will be treated as a [TextFixture] as long as at least one method is
marked with the [Test], [TestCase], or [TestCaseSource] attributes.

[ 148 ]
Unit Testing Chapter 6

The WriteSeparatorLine() method acts as a separator for our debug output. This
method will be called at the top of all our methods in the Tests class:
private static void WriteSeparatorLine()
{
Debug.WriteLine("--------------------------------------------------");
}

The method marked with the [OneTimeSetUp] attribute will only run once before any
tests in that class are run. Any initialization that is required for all the different tests would
be carried out here:
[OneTimeSetUp]
public void OneTimeSetup()
{
WriteSeparatorLine();
Debug.WriteLine("OneTimeSetUp");
Debug.WriteLine("This method is run once before any tests in this
class are run.");
}

The method marked with [OneTimeTearDown] is run once after all the tests have been
run, and before the class is disposed:
[OneTimeTearDown]
public void OneTimeTearDown()
{
WriteSeparatorLine();
Debug.WriteLine("OneTimeTearDown");
Debug.WriteLine("This method is run once after all tests in this
class have been run.");
Debug.WriteLine("This method runs even when an exception occurs.");
}

The method marked with the [Setup] attribute runs once before every test method:
[SetUp]
public void Setup()
{
WriteSeparatorLine();
Debug.WriteLine("Setup");
Debug.WriteLine("This method is run before each test method is run.");
}

[ 149 ]
Unit Testing Chapter 6

The method marked with the [TearDown] attribute is run once after every test method has
completed:
[TearDown]
public void Teardown()
{
WriteSeparatorLine();
Debug.WriteLine("Teardown");
Debug.WriteLine("This method is run after each test method
has been run.");
Debug.WriteLine("This method runs even when an exception occurs.");
}

The Test2() method is a test method as denoted by the [Test] attribute and will be the
second test method to run as determined by the [Order(1)] attribute. This method throws
InconclusiveException:

[Test]
[Order(1)]
public void Test2()
{
WriteSeparatorLine();
Debug.WriteLine("Test:Test2");
Debug.WriteLine("Order: 1");
Assert.Inconclusive("Test 2 is inconclusive.");
}

The Test1() method is a test method as denoted by the [Test] attribute and will be the
first test method to be run as determined by the [0rder(0)] attribute. The method
passes SuccessException:
[Test]
[Order(0)]
public void Test1()
{
WriteSeparatorLine();
Debug.WriteLine("Test:Test1");
Debug.WriteLine("Order: 0");
Assert.Pass("Test 1 passed with flying colours.");
}

The Test3() method is a test method as denoted by the [Test] attribute and will be the
third test method to run as determined by the [Order(2)] attribute. The method
throws AssertionException:
[Test]
[Order(2)]

[ 150 ]
Unit Testing Chapter 6

public void Test3()

{
WriteSeparatorLine();
Debug.WriteLine("Test:Test3");
Debug.WriteLine("Order: 2");
Assert.Fail("Test 1 failed dismally.");
}

When you debug all the tests, your immediate window should look like the following
screenshot:

[ 151 ]
Unit Testing Chapter 6

You have now been exposed to MSTest and NUnit, and have seen the testing life cycle for
each framework in action. It's now time to have a look at Moq.

As you can see from the table of NUnit methods compared to the table of MSTest methods,
NUnit enables more fine-grained unit testing over MSTest and executes with better
performance, which is why it is more widely used than MSTest.

Moq
A unit test should only test the method under test. See the following diagram. If a method
under test calls other methods that are either in the current class or in different classes, then
not only test methods but other methods are also tested:

One way to overcome this is to use mock (fake) objects. The mock object will only test the
method you want to test, and you can make the mock object work any way that you want
to. If you were to write your own mock objects, you would soon come to appreciate that
there is a lot of hard work involved. This may be unacceptable in time-sensitive projects,
and the more complex your code becomes, the more complex your mock objects become.

You will inevitably give it up as a bad job, or you will look for a mocking framework that
suits your needs. Rhino Mocks and Moq are two mocking frameworks for the .NET
Framework. For the purposes of this chapter, we will only be looking at Moq, which is
easier to learn and use compared to Rhino Mocks. For more information on Rhino Mocks,
visit http://hibernatingrhinos.com/oss/rhino-mocks.

When testing using Moq, we start by adding the mock object and then configure the mock
object to do something. We then assert that the configuration is working and that the mock
was invoked. These steps enable us to determine that the mock is correctly set up. Moq only
produces test doubles. It does not test the code. You still need a test framework such as
NUnit to test your code.

[ 152 ]
Unit Testing Chapter 6

We'll now look at an example of using Moq and NUnit together.

Create a new console application and call it CH06_Moq. Add the following interface and
classes—IFoo, Bar, Baz, and UnitTests. Then, via the Nuget package manager, install
Moq, NUnit, and NUnit3TestAdapter. Update the Bar class with the following code:
namespace CH06_Moq
{
public class Bar
{
public virtual Baz Baz { get; set; }
public virtual bool Submit() { return false; }
}
}

The Bar class has a virtual property of type Baz and a virtual method called Submit() that
returns a Boolean value of false. Now update the Baz class as follows:
namespace CH06_Moq
{
public class Baz
{
public virtual string Name { get; set; }
}
}

The Baz class has a single virtual property of type string called Name. Modify the IFoo file
to contain the following source code:
namespace CH06_Moq
{
public interface IFoo
{
Bar Bar { get; set; }
string Name { get; set; }
int Value { get; set; }
bool DoSomething(string value);
bool DoSomething(int number, string value);
string DoSomethingStringy(string value);
bool TryParse(string value, out string outputValue);
bool Submit(ref Bar bar);
int GetCount();
bool Add(int value);
}
}

[ 153 ]
Unit Testing Chapter 6

The IFoo interface has a number of properties and methods. As you can see, the interface
has a reference to the Bar class, and we know that the Bar class contains a reference to the
Baz class. We will now start updating our UnitTests class to test our newly-created
interface and classes using NUnit and Moq. Modify the UnitTests class file so that it looks
like the code that follows:
using Moq;
using NUnit.Framework;
using System;

namespace CH06_Moq
{
[TestFixture]
public class UnitTests
{
}
}

Now, add the AssertThrows method that asserts whether a designated exception has been
thrown or not:
public bool AssertThrows<TException>(
Action action,
Func<TException, bool> exceptionCondition = null
) where TException : Exception
{
try
{
action();
}
catch (TException ex)
{
if (exceptionCondition != null)
{
return exceptionCondition(ex);
}
return true;
}
catch
{
return false;
}
return false;
}

[ 154 ]
Unit Testing Chapter 6

The AssertThrows method is a generic method that will return true if your method
throws the designated exception, and false if it does not. We will be using this method
when we test exceptions further in this chapter. Now, add the
DoSomethingReturnsTrue() method:

[Test]
public void DoSomethingReturnsTrue()
{
var mock = new Mock<IFoo>();
mock.Setup(foo => foo.DoSomething("ping")).Returns(true);
Assert.IsTrue(mock.Object.DoSomething("ping"));
}

The DoSomethingReturnsTrue() method creates a new mock implementation of the

IFoo interface. Then it sets up the DoSomething() method to accept a string containing
the word "ping", and then returns true. Finally, the method asserts that when the
DoSomething() method is called with the text "ping", the method returns a value of
true. We'll now implement a similar test method that returns false if the value is
"tracert":

[Test]
public void DoSomethingReturnsFalse()
{
var mock = new Mock<IFoo>();
mock.Setup(foo => foo.DoSomething("tracert")).Returns(false);
Assert.IsFalse(mock.Object.DoSomething("tracert"));
}

The DoSomethingReturnsFalse() method follows the same procedure as the

DoSomethingReturnsFalse() method. We create a mock object of the IFoo interface, set
it up to return false if the parameter value is "tracert", and then assert that false is
returned for a parameter value of "tracert". Next, we'll test our arguments:
[Test]
public void OutArguments()
{
var mock = new Mock<IFoo>();
var outString = "ack";
mock.Setup(foo => foo.TryParse("ping", out outString)).Returns(true);
Assert.AreEqual("ack", outString);
Assert.IsTrue(mock.Object.TryParse("ping", out outString));
}

[ 155 ]
Unit Testing Chapter 6

The OutArguments() method creates an implementation of the IFoo interface. A string

that will be used as an out parameter is then declared and assigned the value "ack". Next,
the TryParse() method of the IFoo mock object is set up to return true for an input value
of "ping" and to output the string value of "ack". We then assert that the outString is
equal to the value "ack". The final check asserts that TryParse() returns true for the
input value of "ping":
[Test]
public void RefArguments()
{
var instance = new Bar();
var mock = new Mock<IFoo>();
mock.Setup(foo => foo.Submit(ref instance)).Returns(true);
Assert.AreEqual(true, mock.Object.Submit(ref instance));
}

The RefArguments() method creates an instance of the Bar class. Then, a mock
implementation of the IFoo interface is created. The Submit() method is then set up to
return true if the reference type passed in is of type Bar. We then assert that the argument
that is passed in is true of type Bar. In our AccessInvocationArguments() test method,
we create a new implementation of the IFoo interface:
[Test]
public void AccessInvocationArguments()
{
var mock = new Mock<IFoo>();
mock.Setup(foo => foo.DoSomethingStringy(It.IsAny<string>()))
.Returns((string s) => s.ToLower());
Assert.AreEqual("i like oranges!", mock.Object.DoSomethingStringy("I
LIKE ORANGES!"));
}

Then we set up the DoSomethingStringy() method to convert the input to lowercase and
return it. Finally, we assert that the string returned is the string passed in that has been
converted to lowercase:
[Test]
public void ThrowingWhenInvokedWithSpecificParameters()
{
var mock = new Mock<IFoo>();
mock.Setup(foo => foo.DoSomething("reset"))
.Throws<InvalidOperationException>();
mock.Setup(foo => foo.DoSomething(""))
.Throws(new ArgumentException("command"));
Assert.IsTrue(
AssertThrows<InvalidOperationException>(

[ 156 ]
Unit Testing Chapter 6

() => mock.Object.DoSomething("reset")
)
);
Assert.IsTrue(
AssertThrows<ArgumentException>(
() => mock.Object.DoSomething("")
)
);
Assert.Throws(
Is.TypeOf<ArgumentException>()
.And.Message.EqualTo("command"),
() => mock.Object.DoSomething("")
);
}

In our final test method called ThrowingWhenInvokedWithSpecificParameters(), we

create a mock implementation of the IFoo interface. We then configure the
DoSomething() method to throw InvalidOperationException when the passed-in
value is "reset".

An ArgumentException exception of "command" is thrown when an empty string is

passed in. We then assert that InvalidOperationException is thrown when the input
value is "reset". When the input value is an empty string, we assert that
ArgumentException is thrown, with the assertion that the message of
ArgumentException is "command".

You've now seen how to use a mocking framework called Moq to create mock objects to
test your code using NUnit. The last tool we will now look at is called SpecFlow. SpecFlow
is a BDD tool.

SpecFlow
User-focused behavioral tests that are written ahead of the code are the primary function
behind BDD. BDD is a software development methodology that evolved from TDD. You
start BDD with a list of features. Features are specifications written in a formal business
language. This language is understandable by all stakeholders on a project. Once the
features have been agreed and generated, it is up to the developers to then develop step
definitions for the feature statements. Once the step definitions have been created, the next
step is to create the external project to implement the feature and add a reference to it. The
step definitions are then extended to implement the application code for the feature.

[ 157 ]
Unit Testing Chapter 6

One benefit of this approach is that you, as a programmer, are guaranteed to deliver on
what the business has asked for, rather than give them what you think they asked for. This
can save the business a lot of money and hours. Past history has shown that many projects
failed because of the lack of clarity on what needed to be delivered between the business
teams and the programming teams. BDD helps to alleviate this potential hazard when
developing new features.

In this section of the chapter, we will develop a very simple calculator example using the
BDD software development methodology by using SpecFlow.

We will start by writing a feature file that will act as our specification with acceptance
criteria. Then we will generate our step definitions from our feature file that will generate
our required methods. Once our step definitions have generated the required methods, we
will then write the code for them so that our feature is complete.

Create a new class library and add the following packages—NUnit, NUnit3TestAdapter,
SpecFlow, SpecRun.SpecFlow and SpecFlow.NUnit. Add a new SpecFlow Feature file
called Calculator:
Feature: Calculator
In order to avoid silly mistakes
As a math idiot
I want to be told the sum of two numbers

@mytag
Scenario: Add two numbers
Given I have entered 50 into the calculator
And I have entered 70 into the calculator
When I press add
Then the result should be 120 on the screen

The preceding text is the text automatically added to the Calculator.feature file upon
creation. So we'll use this as our starting point for learning BDD using SpecFlow. As of the
time of writing, it worth noting that SpecFlow and SpecMap have been acquired by
Tricentis. Tricentis has stated that SpecFlow, SpecFlow+, and SpecMap will all remain free,
so now is a good time to learn and use SpecFlow and SpecMap if you haven't already done
so.

[ 158 ]
Unit Testing Chapter 6

Now that we have our feature file, we need to create step definitions that will bind our
feature request to our code. Right-click in the code editor and a context menu will pop up.
Select Generate step definitions. You should see the following dialog:

Enter the name CalculatorSteps for the class name. Click on the Generate button to
generate the step definition and save the file. Open the CalculatorSteps.cs file and you
should see the following code:
using TechTalk.SpecFlow;

namespace CH06_SpecFlow
{
[Binding]
public class CalculatorSteps
{
[Given(@"I have entered (.*) into the calculator")]
public void GivenIHaveEnteredIntoTheCalculator(int p0)
{
ScenarioContext.Current.Pending();
}

[ 159 ]
Unit Testing Chapter 6

[When(@"I press add")]

public void WhenIPressAdd()
{
ScenarioContext.Current.Pending();
}
[Then(@"the result should be (.*) on the screen")]
public void ThenTheResultShouldBeOnTheScreen(int p0)
{
ScenarioContext.Current.Pending();
}
}
}

A comparison of the contents of the steps file with the feature file is shown in the following
screenshot:

The code that implements the feature must be in a separate file. Create a new class library
and call it CH06_SpecFlow.Implementation. Then, add a file called Calculator.cs.
Add a reference to the newly created library in the SpecFlow project, and the following line
to the top of the CalculatorSteps.cs file:
private Calculator _calculator = new Calculator();

[ 160 ]
Unit Testing Chapter 6

We are now in a position to extend our step definitions so that they implement the
application code. In the CalculatorSteps.cs file, replace all the p0 parameters with a
number. This makes the parameter requirement more explicit. At the top of the Calculate
class, add two public properties called FirstNumber and SecondNumber, as shown in the
code that follows:
public int FirstNumber { get; set; }
public int SecondNumber { get; set; }

In the CalculatorSteps class, update the GivenIHaveEnteredIntoTheCalculator()

method as shown:
[Given(@"I have entered (.*) into the calculator")]
public void GivenIHaveEnteredIntoTheCalculator(int number)
{
calculator.FirstNumber = number;
}

Now, add the second method, GivenIHaveAlsoEnteredIntoTheCalculator(), if it

does not already exist, and assign the number parameter to the calculator's second number:
public void GivenIHaveAlsoEnteredIntoTheCalculator(int number)
{
calculator.SecondNumber = number;
}

Add private int result; to the top of the CalculatorSteps class and before any
steps. Add the Add() method to the Calculator class:
public int Add()
{
return FirstNumber + SecondNumber;
}

Now, update the WhenIPressAdd() method in the CalculatorSteps class and update
the result variable with the result of calling the Add() method:
[When(@"I press add")]
public void WhenIPressAdd()
{
_result = _calculator.Add();
}

Next up, modify the ThenTheResultShouldBeOnTheScreen() method as follows:

[Then(@"the result should be (.*) on the screen")]
public void ThenTheResultShouldBeOnTheScreen(int expectedResult)

[ 161 ]
Unit Testing Chapter 6

{
Assert.AreEqual(expectedResult, _result);
}

Build your project and run your tests. You should see that the tests pass. Only the code
required by the feature to pass has been written and your code has passed the test.

You can find out more about SpecFlow at https://specflow.org/docs/. We've covered
some of the tools available for you to develop and test your code. Now it is time to see a
really simple example of how we go about coding using TDD. We'll start by writing code
that fails. Then, we'll write just enough code for the test to compile. And finally, we will
refactor the code.

TDD methodology practice – fail, pass, and

refactor
In this section, you will learn to write tests that fail. Then you will learn to write just
enough code to make the test pass, and then if necessary, you will perform any refactoring
that needs to take place.

Before we delve into a practical example of TDD, let's consider why we need TDD. In the
previous section, you saw how we can create feature files and generate step files from them
to write code that meets a business need. Another way to ensure that your code meets the
business requirements is with TDD. With TDD, you start with a test that fails. Then, you
write just enough code to make the test pass, and as the need arises, you perform
refactoring of your new code. This process is repeated until such time as all the features
have been coded.

But why do we need TDD?

Business software specifications are put together by business analysts who work with
project stakeholders to design new software, or extensions and modifications to existing
software. Some software is critical and cannot afford to be buggy. Such software includes
financial systems that handle private and business investments; medical equipment,
including critical life support and scanning equipment, that requires functional software for
it to work; transport signaling software for traffic management and navigation systems;
space flight systems; and weapon systems.

Okay, but where does TDD fit in?

[ 162 ]
Unit Testing Chapter 6

Well, you've been given a specification to write a piece of software. The first thing you need
to do is create your project. Then, you write the pseudocode for the functionality that you
are going to implement. You then progress to writing the tests for each piece of
pseudocode. The test fails. You then write the required code that causes the test to pass, and
then you refactor your code as needed. What you are doing here is writing code that is well
tested and robust. You are able to guarantee that your code will execute as expected in
isolation. If your code is a component of a larger system, then it will be the responsibility of
the test team to test the integration of your code, not you. You, as a developer, have earned
the confidence in your code to release it to the test team. If the test team identify use cases
that have previously been overlooked, they will share them with you. You will then write
further tests and make them pass before releasing the updated code to them. Such a way of
working ensures that code is of the highest standard and can be trusted to work as expected
by given the expected outputs for the given inputs. And finally, TDD makes software
progress measurable, which is good news for managers.

It's time for our little demonstration of TDD. In this example, we will use TDD to develop a
simple logging application that can handle inner exceptions, and logs exceptions to a
timestamped text file. We will write the program and get the tests to pass. Once we have
written our program and got all the tests to pass, then we will refactor our code to make it
reusable and easier to read, and of course, we will make sure that our tests still pass.

1. Create a new console application and call it CH06_FailPassRefactor. Add a

class called UnitTests with the following pseudocode:
using NUnit.Framework;

namespace CH06_FailPassRefactor
{
[TestFixture]
public class UnitTests
{
// The PseudoCode.
// [1] Call a method to log an exception.
// [2] Build up the text to log including
// all inner exceptions.
// [3] Write the text to a file with a timestamp.
}
}

[ 163 ]
Unit Testing Chapter 6

2. We'll write our first unit test to satisfy the condition [1]. In our unit test, we will
test create the Logger variable, call the Log() method, and pass the test. So, let's
write the code:
// [1] Call a method to log an exception.
[Test]
public void LogException()
{
var logger = new Logger();
var logFileName = logger.Log(new ArgumentException("Argument
cannot be null"));
Assert.Pass();
}

This test will not run as the project will not build. That is because the Logger
class does not exist. So add an internal class called Logger to the project. Then
run your test. The build will still fail, and the test won't be run because we are
now missing the Log() method. So let's add the Log() method to our Logger
class. Then, we'll try and run our test again. This time, the test should succeed.

3. At this stage, we will perform any necessary refactoring. But since we have just
started, there is no refactoring to do, so we can move on to our next test.

Our code to generate the log message and save it to disk will feature private
members. With NUnit, you don't test private members. The school of thought is
that if you have to test private members, then there must be something wrong
with your code. So, we'll move on to our next unit test, which will determine
whether the log file exists. Before we write our unit test, we will write a method
that returns an exception with an inner exception that has an inner exception. We
will pass the returned exception into the Log() method in our unit test:
private Exception GetException()
{
return new Exception(
"Exception: Main exception.",
new Exception(
"Exception: Inner Exception.",
new Exception("Exception: Inner Exception Inner
Exception")
)
);
}

[ 164 ]
Unit Testing Chapter 6

4. Now, we have our GetException() method in place where we can write our
unit test to check whether the log file exists:
[Test]
public void CheckFileExists()
{
var logger = new Logger();
var logFile = logger.Log(GetException());
FileAssert.Exists(logFile);
}

5. If we build our code and run the CheckFileExists() test, it will fail, so we
need to write the code for it to succeed. In the Logger class, add private
StringBuilder _stringBuilder; to the top of the Logger class. Then,
modify the Log() method and add the following method to the Logger class:
private StringBuilder _stringBuilder;

public string Log(Exception ex)

{
_stringBuilder = new StringBuilder();
return SaveLog();
}

private string SaveLog()

{
var fileName = $"LogFile{DateTime.UtcNow.GetHashCode()}.txt";
var dir =
Environment.GetFolderPath(Environment.SpecialFolder.MyDocuments);
var file = $"{dir}\\{fileName}";
return file;
}

6. We have called the Log() method and a log file is generated. Now, all we need is
the text to be logged to the file. According to our pseudocode, we need to log the
main exception and all inner exceptions. Let's write a test that checks whether the
log file contains the message "Exception: Inner Exception Inner
Exception":

[Test]
public void ContainsMessage()
{
var logger = new Logger();
var logFile = logger.Log(GetException());
var msg = File.ReadAllText(logFile);
Assert.IsTrue(msg.Contains("Exception: Inner Exception Inner

[ 165 ]
Unit Testing Chapter 6

Exception"));
}

7. Now, we know that the test will fail because the string builder is empty, so we
will add the method to the Logger class that will take an exception, log the
message, and check whether the exception has an inner exception. If it has, then
it will call itself with the parameter isInnerException:
private void BuildExceptionMessage(Exception ex, bool
isInnerException)
{
if (isInnerException)
_stringBuilder.Append("Inner Exception:
").AppendLine(ex.Message);
else
_stringBuilder.Append("Exception:
").AppendLine(ex.Message);
if (ex.InnerException != null)
BuildExceptionMessage(ex.InnerException, true);
}

8. Finally, update the Log() method of the Logger class to call our
BuildExceptionMessage() method:

public string Log(Exception ex)

{
_stringBuilder = new StringBuilder();
_stringBuilder.AppendLine("-----------------------
-----------------");
BuildExceptionMessage(ex, false);
_stringBuilder.AppendLine("-----------------------
-----------------");
return SaveLog();
}

All our tests now pass and we have a fully functioning program that does what's expected
of it, but there is an opportunity here for some refactoring. The method called
BuildExceptionMessage() is a candidate for reuse as it is very useful for debugging
purposes, especially when you have an exception with an inner exception, so we are going
to move that method into its own method. Notice that the Log() method is also building
the opening and closing portions of the text to be logged.

[ 166 ]
Unit Testing Chapter 6

We can and will move this into the BuildExceptionMessage() method:

1. Create a new class and call it Text. Add a private StringBuilder member
variable and instantiate it in the constructor. Then, update the class by adding the
following code:
public string ExceptionMessage => _stringBuilder.ToString();

public void BuildExceptionMessage(Exception ex, bool

isInnerException)
{
if (isInnerException)
{
_stringBuilder.Append("Inner Exception:
").AppendLine(ex.Message);
}
else
{
_stringBuilder.AppendLine("--------------------------------
------------------------------");
_stringBuilder.Append("Exception:
").AppendLine(ex.Message);
}
if (ex.InnerException != null)
BuildExceptionMessage(ex.InnerException, true);
else
_stringBuilder.AppendLine("--------------------------------
------------------------------");
}

2. We've now got a useful Text class that returns a useful exception message from
an exception with inner exceptions, but we can also refactor the code in the
SaveLog() method. We can extract the code that generates a unique hashed
filename into its own method. So, let's add the following method to the Text
class:
public string GetHashedTextFileName(string name, SpecialFolder
folder)
{
var fileName = $"{name}-{DateTime.UtcNow.GetHashCode()}.txt";
var dir = Environment.GetFolderPath(folder);
return $"{dir}\\{fileName}";
}

[ 167 ]
Unit Testing Chapter 6

3. The GetHashedTextFileName() method accepts a name for the file specified by

the user and a special folder. It then adds a hyphen and the current UTC date's
hash code to the end of the filename. It then adds the .txt file extension and
assigns the text to the fileName variable. The absolute path of the special folder
requested by the caller is then assigned to the dir variable and the path and
filename are then returned to the user. This method is guaranteed to return
unique filenames.
4. Replace the body of the Logger class with the following code:
private Text _text;

public string Log(Exception ex)

{
BuildMessage(ex);
return SaveLog();
}

private void BuildMessage(Exception ex)

{
_text = new Text();
_text.BuildExceptionMessage(ex, false);
}

private string SaveLog()

{
var filename = _text.GetHashedTextFileName("Log",
Environment.SpecialFolder.MyDocuments);
File.WriteAllText(filename, _text.ExceptionMessage);
return filename;
}

[ 168 ]
Unit Testing Chapter 6

The class is still doing the same thing, but it is cleaner and smaller as the message and
filename generation has been moved to a separate class. If you run the code, it behaves in
the same way. If you run the tests, they will all pass.

In this section, we have written unit tests that failed, and then modified them so that they
passed. Then, we refactored the code to make it cleaner, which resulted in us writing code
that can be reused in the same project or other projects. Let's now take a very brief look at
redundant tests.

Removing redundant tests, comments, and

dead code
As the book states, we are interested in writing clean code. As our programs and tests grow
and we start to refactor, some code will become redundant. Any code that is redundant and
does not get called is known as dead code. Dead code should always be removed as soon as
it is identified. Dead code will not be executed in compiled code, but it is still part of the
code base that needs to be maintained. Code files with dead code are longer than they need
to be. Apart from the unnecessary fact that it makes your files bigger, it can also make
reading source code harder, as it may cut through the natural flow of the code and add
confusion and delay to the programmer reading it. Not only that, but the last thing any
programmer new to the project needs is to waste valuable time trying to understand dead
code that will never be used. So it is best to get rid of it.

As for comments, they can be really useful if done right, and API commenting is
particularly beneficial for API documentation generation. But some comments just add
noise to the code file, and a surprising number of programmers can become really irritated
by them. There is one group of programmers that will comment on everything. Another
group won't comment on anything as they believe the code should read like a book. And
then there are those who take a balanced approach, and only comment on code when it is
deemed necessary for people to understand the code.

[ 169 ]
Unit Testing Chapter 6

When you see comments like this—"This generates a random bug every so often. Don't know
why. But you're welcome to fix it!"—alarm bells should start ringing. First of all, the
programmer who wrote the comment should have stuck with the code and not moved on
until the conditions that generate the bug were identified, and then the bug should have
been fixed. If you know who the programmer is who wrote the comment, then return the
code to them to fix and remove the comment. I have seen code like that on more than one
occasion, and I've seen comments on the web expressing these strong sentiments about
such comments. I suppose it is a way to deal with lazy programmers. Should they not be
lazy, but rather simply inexperienced, then it is a good learning task in the art of problem
diagnosis and resolution.

If code has been checked in and approved, and you come across blocks of code that have
been commented out, then delete them. The code will still exist in the version control
history and you will be able to retrieve it from there if you have to.

Code should be read like a book, and so you should not aim to make your code cryptic just
to look good and impress your colleagues, because I guarantee that when you come back to
your own code in a few weeks' time, you will scratch your head wondering what your own
code does and why. I've seen many juniors make this mistake.

Redundant tests should also be removed. You only need to run the tests that are necessary.
Tests for redundant code have no value and can waste considerable time. Also, if your
company has CI/CD pipelines that also run tests in the cloud, then the redundant tests and
dead code add business costs to the build, test, and deploy pipelines. This means that the
fewer lines of code you upload, build, test, and deploy, the less your company has to fork
out on running costs. Remember, running processes in the cloud costs money and the aim
of a business is to spend as little money as possible, but rake in plenty of money.

So now that we've finished the chapter, let's summarize what we've learned.

[ 170 ]
Unit Testing Chapter 6

Summary
We started by looking at why it is important for developers to write unit tests to develop
quality-assured code. Theoretical problems were identified that could arise from bugs in
the software. These include loss of life and expensive lawsuits. Unit testing and what makes
a good unit test was then discussed. We identified that a good unit test must be atomic,
deterministic, repeatable, and fast.

Next, we went on to look at the tools available to developers that assist with TDD and BDD.
MSTest and NUnit were discussed with examples that showed how to implement TDD.
Then, we looked at using a mocking framework called Moq in conjunction with NUnit for
testing mock objects. Our look at tools then concluded with SpecFlow—a BDD tool that
allows us to write features in a business language that both techies and non-techies can
understand—to make sure that what the business wants is what the business gets.

NUnit was then put to work as we worked through a very simple TDD example using the
fail, pass, and refactor methodology, before finally looking at why we should remove
unnecessary comments, redundant tests, and dead code.

At the end of this chapter, you will find further resources on testing software programs. In
the next chapter, we are going to look at end-to-end testing. But before that, you might as
well have a go at the following questions and see how much knowledge on unit testing you
have retained.

Questions
1. What makes a good unit test?
2. What should a good unit test not be?
3. What does TDD stand for?
4. What does BDD stand for?
5. What is a unit test?
6. What is a mock object?
7. What is a fake object?
8. Name some unit testing frameworks.
9. Name some mocking frameworks.
10. Name a BDD framework.
11. What should be removed from source code files?

[ 171 ]
Unit Testing Chapter 6

Further reading
A brief overview of unit testing, with links to further information on different
types of unit testing including integration testing, acceptance testing, and tester
job descriptions, can be found at http://softwaretestingfundamentals.com/
unit-testing.
The Rhino Mocks homepage can be found at http://hibernatingrhinos.com/
oss/rhino-mocks.

[ 172 ]
7
End-to-End System Testing
End-to-end (E2E) system testing is the automated testing of a system in its entirety. As a
programmer, the unit tests for your piece of code are just a small factor in the bigger picture
of the whole system. So in this chapter, we will be looking at the following topics:

Performing E2E testing

Coding and testing factories
Coding and testing dependency injection
Testing modularization

By the end of this chapter, you will have gained the following skills:

Be able to define E2E testing

Be able to perform E2E testing
Be able to explain what factories are and how to use them
Be able to understand what dependency injection is and how to use it
Be able to understand what modularization is and how to utilize it

E2E testing
So, you've finished your project and all the unit tests pass. However, your project is a part
of a larger system. This larger system will need to be tested to make sure that your code,
and the other code it interfaces with, both work together as expected. Code tested in
isolation can break when integrated into larger systems, and existing systems can break
with the addition of new code, so it is important to perform E2E testing, also known as
integration testing.
End-to-End System Testing Chapter 7

Integration testing is responsible for testing the complete program flow from beginning to
end. Integration testing usually starts at the requirements gathering stage. You start by
gathering and documenting the various requirements of the system. You then design all the
components and devise tests for each subsystem, and then the E2E tests for the whole
system. Then, you write your code according to the requirements and implement your own
unit tests. Once your code is complete and the tests all pass, then the code is integrated into
the overall system within the test environment and the E2E tests are executed. Often, E2E
tests are carried out manually, although where possible, they can be automated as well. The
following diagram shows a system that comprises two subsystems with modules and a
database. In E2E testing, all these modules will be tested either manually, using
automation, or by both methods:

The input to and output from each system are the main focus of the tests. You have to ask
yourself, is the correct information passed in and passed out of each system?

Additionally, there are three things to consider when building your E2E tests:

What user functions will there be, and what steps will each function perform?
What conditions will there be for each function and each of its steps?
What are the different scenarios that we will have to build test cases for?

Each subsystem will have one or more features that it will provide, and each feature will
have a number of actions that will be executed in a particular order. Those actions will
receive inputs and provide outputs. There will also be relationships between features and
functions that you must identify, after which you will need to determine whether the
function is reusable or independent.

[ 174 ]
End-to-End System Testing Chapter 7

Consider the scenario of an online testing product. Teachers and students will log in to the
system. If the teacher logs in, they will be taken to an admin console, and if a student logs
in, they will be taken to the test menu to carry out one or more tests. In this scenario, we
effectively have three subsystems:

The login system

The admin system
The test system

There are two flows of execution in the aforementioned system. We have the admin flow
and the test flow. Conditions and test cases will have to be established for each flow. We
will use this very simple assessment system login scenario for our E2E example. In the real
world, E2E will be more involved than in this chapter. The main aim of this chapter is to get
you thinking about E2E testing and how you can best implement it, so we will keep things
as simple as we can so that complexity does not get in the way of what we are trying to
accomplish, which is to manually test three modules that must interact with each other.

The aim of this section is to build three console applications that make up the complete
system: the login module, the admin module, and the test module. Then once they are built,
we will go through testing them manually. The diagram that follows displays the
interaction between systems. We will start with the login module:

The login module (subsystem)

The first part of our system requires both teachers and students to log in to the system
using a username and password. The task list is as follows:

1. Enter the username.

2. Enter the password.
3. Press Cancel (this resets username and password).
4. Press OK.

[ 175 ]
End-to-End System Testing Chapter 7

5. If the username is invalid, then display an error message on the login page.
6. If the user is valid, then do the following:
If the user is a teacher, load the admin console.
If the user is a student, load the test console.

Let's start by creating a console application. Call it CH07_Logon. In the Program.cs class,
replace the existing code with the following:
using System;
using System.Collections.Generic;
using System.Diagnostics;
using System.Linq;

namespace CH07_Logon
{
internal static class Program
{
private static void Main(string[] args)
{
DoLogin("Welcome to the test platform");
}
}
}

The DoLogin() method will take the passed-in string and use it for the title. Since we will
not have logged in yet, the title will be set to "Welcome to the test platform". We
need to add the DoLogin() method. The code for this method is as follows:
private static void DoLogin(string message)
{
Console.WriteLine("----------------------------");
Console.WriteLine(message);
Console.WriteLine("----------------------------");
Console.Write("Enter your username: ");
var usr = Console.ReadLine();
Console.Write("Enter your password: ");
var pwd = ReadPassword();
ValidateUser(usr, pwd);
}

The previous code accepts a message. The message is used as the title in the console
window. The user is then prompted to enter their username and password. The
ReadPassword() method reads all inputs and replaces filtered letters with an asterisk to
hide the user's input. The username and password are then validated by calling the
ValidateUser() method.

[ 176 ]
End-to-End System Testing Chapter 7

The next thing we must do is add the ReadPassword() method as in the code that follows:
public static string ReadPassword()
{
return ReadPassword('*');
}

This method is really simple. It calls an overloaded method of the same name and passes in
the password mask character. Let's implement the overloaded ReadPassword() method:
public static string ReadPassword(char mask)
{
const int enter = 13, backspace = 8, controlBackspace = 127;
int[] filtered = { 0, 27, 9, 10, 32 };
var pass = new Stack<char>();
char chr = (char)0;
while ((chr = Console.ReadKey(true).KeyChar) != enter)
{
if (chr == backspace)
{
if (pass.Count > 0)
{
Console.Write("\b \b");
pass.Pop();
}
}
else if (chr == controlBackspace)
{
while (pass.Count > 0)
{
Console.Write("\b \b");
pass.Pop();
}
}
else if (filtered.Count(x => chr == x) <= 0)
{
pass.Push((char)chr);
Console.Write(mask);
}
}
Console.WriteLine();
return new string(pass.Reverse().ToArray());
}

[ 177 ]
End-to-End System Testing Chapter 7

The overloaded ReadPassword() method accepts a password mask. This method adds
each character to the stack. Unless the key being pressed is the Enter key, the key being
pressed is checked to see if the user is performing a Delete keypress. If the user is
performing a Delete keypress, then the last character entered is removed from the stack. If
the character entered is not in the filtered list, then it is pushed onto the stack. The
password mask is then written to the screen. As soon as the Enter key is pressed, a blank
line is written to the console window, and the contents of the stack are reversed, returning
it as a string.

The final method we need to write for this subsystem is the ValidateUser() method:
private static void ValidateUser(string usr, string pwd)
{
if (usr.Equals("admin") && pwd.Equals("letmein"))
{
var process = new Process();
process.StartInfo.FileName =
@"..\..\..\CH07_Admin\bin\Debug\CH07_Admin.exe";
process.StartInfo.Arguments = "admin";
process.Start();
}
else if (usr.Equals("student") && pwd.Equals("letmein"))
{
var process = new Process();
process.StartInfo.FileName =
@"..\..\..\CH07_Test\bin\Debug\CH07_Test.exe";
process.StartInfo.Arguments = "test";
process.Start();
}
else
{
Console.Clear();
DoLogin("Invalid username or password");
}
}

The ValidateUser() method checks the username and password. If they validate as an
admin, then the admin page is loaded. If they validate as a student, then the student page is
loaded. Otherwise, the console is cleared, the user is informed the credentials are wrong,
and they are prompted to reenter their credentials.

[ 178 ]
End-to-End System Testing Chapter 7

Upon a successful login operation being performed, the relevant subsystem is loaded and
the login subsystem then terminates. Now that we have written our login module, we will
write our admin module.

The admin module (subsystem)

The admin subsystem is where all the system administration is carried out. This includes
the following:

Importing students
Exporting students
Adding students
Deleting students
Editing students' profiles
Assigning tests to students
Changing the administrator password
Backing up data
Restoring data
Erasing all data
Viewing reports
Exporting reports
Saving reports
Printing reports
Logging out

For this exercise, we will not be implementing any of these features. I will leave you to do
that as a fun exercise. All we are interested in is that the admin module loads on a
successful login. If the admin module is loaded without logging in, then an error message is
displayed. Then when the user presses a key, they are taken to the login module. Successful
login is accomplished when a user successfully logs in as an administrator, and the admin
executable is called with the admin argument.

Create a console application in Visual Studio and call it CH07_Admin. Update the Main()
method as follows:
private static void Main(string[] args)
{
if ((args.Count() > 0) && (args[0].Equals("admin")))
{
DisplayMainScreen();

[ 179 ]
End-to-End System Testing Chapter 7

}
else
{
DisplayMainScreenError();
}
}

The Main() method checks that the argument count is greater than 0 and that the first
argument in the array is admin. If it is, then the main screen is displayed by calling the
DisplayMainScreen() method. Otherwise, the DisplayMainScreenError() method is
called that warns the user that they must log in to access the system. It's time to write the
DisplayMainScreen() method:

private static void DisplayMainScreen()

{
Console.WriteLine("------------------------------------");
Console.WriteLine("Test Platform Administrator Console");
Console.WriteLine("------------------------------------");
Console.WriteLine("Press any key to exit");
Console.ReadKey();
Process.Start(@"..\..\..\CH07_Logon\bin\Debug\CH07_Logon.exe");
}

As you can see, the DisplayMainScreen() method is really simple. It displays a title with
a message to press any key to exit, then waits for a keypress. Upon keypress, the program
shells out to the login module and exits. Now, for the DisplayMainScreenError()
method:
private static void DisplayMainScreenError()
{
Console.WriteLine("------------------------------------");
Console.WriteLine("Test Platform Administrator Console");
Console.WriteLine("------------------------------------");
Console.WriteLine("You must login to use the admin module.");
Console.WriteLine("Press any key to exit");
Console.ReadKey();
Process.Start(@"..\..\..\CH07_Logon\bin\Debug\CH07_Logon.exe");
}

From this method, you can see that the module was started without logging in. This is not
permitted. So when the user presses any key, the user is redirected to the login module,
where they can log in to use the admin module. Our final module is the test module. Let's
get to work and write it.

[ 180 ]
End-to-End System Testing Chapter 7

The test module (subsystem)

The test system consists of a menu. This menu displays a list of tests the student must
perform, and also provides the option to exit the test system. The functions of this system
include the following:

Display a menu of tests to be completed.

From the menu, select an item to start a test.
On test completion, save results and return to the menu.
When a test has been completed, remove it from the menu.
When the user exits the test module, they are returned to the login module.

As with the previous module, I will let you have a play and add the aforementioned
functionality. The main thing we are interested in here is to make sure the test module can
only be run when the user has logged in. When the module is exited, the login module is
loaded.

The test module is more or less a rehash of the admin module, so we will rush through this
section to get to where we need to be. Update the Main() method as follows:
private static void Main(string[] args)
{
if ((args.Count() > 0) && (args[0].Equals("test")))
{
DisplayMainScreen();
}
else
{
DisplayMainScreenError();
}
}

Now add the DisplayMainScreen() method:

private static void DisplayMainScreen()
{
Console.WriteLine("------------------------------------");
Console.WriteLine("Test Platform Student Console");
Console.WriteLine("------------------------------------");
Console.WriteLine("Press any key to exit");
Console.ReadKey();
Process.Start(@"..\..\..\CH07_Logon\bin\Debug\CH07_Logon.exe");
}

[ 181 ]
End-to-End System Testing Chapter 7

And finally, write the DisplayMainScreenError() method:

private static void DisplayMainScreenError()
{
Console.WriteLine("------------------------------------");
Console.WriteLine("Test Platform Student Console");
Console.WriteLine("------------------------------------");
Console.WriteLine("You must login to use the student module.");
Console.WriteLine("Press any key to exit");
Console.ReadKey();
Process.Start(@"..\..\..\CH07_Logon\bin\Debug\CH07_Logon.exe");
}

Now that we have written all three modules, we will test them in the next section.

Testing our three-module system using E2E

In this section, we are going to perform a manual E2E test of our three-module system. We
will test the login module to ensure that it only allows valid logins access to either the
admin module or the test module. When a valid admin logs into the system, they should
see the admin module, and the login module should be unloaded. When a valid student
logs into the system, then they should see the test module, and the login module should be
unloaded.

If we then try and load the admin module without first logging in, we should be warned
that we must log in. Pressing any key should unload the admin module and load the login
module. Trying to use the test module without logging in should behave in the same way
as the admin module. We should be warned that we can't use the test module unless we log
in, and pressing any key should load the login module and unload the test module.

Let's now go through the manual testing process:

1. Make sure that all the projects are built, then run the login module. You should
see the screen that follows:

[ 182 ]
End-to-End System Testing Chapter 7

2. Enter an incorrect username and/or password, then press Enter, and you will see
the following screen:

3. Now, enter admin as the username and letmein as the password, and then
press Enter. You should see the admin module screen for a successful login:

4. Press any key to exit, and you should see the login module again:

[ 183 ]
End-to-End System Testing Chapter 7

5. Enter student as your username and letmein as your password. Press Enter
and you should be shown the student module:

6. Now load the admin module without logging in, and you should see the
following:

7. Pressing any key will take you back to the login module. Now load the test
module without logging in, and you should see the following:

[ 184 ]
End-to-End System Testing Chapter 7

We have now successfully manually carried out E2E testing of our system that consists of
three modules. This is by far the best way to run through a system when E2E testing. Your
unit tests will be very useful in making this stage fairly straightforward. By the time you get
to this stage, your bugs should have been caught and dealt with. But as always, there is
always the possibility of problems being encountered, which is why it is good to manually
run through the system as a whole manually. That way, you can visually see through your
interactions that the system behaves as expected.

Larger systems employ factories and dependency injection. In the following sections of this
chapter, we will look at them both, starting with factories.

Factories
Factories are implemented using the factory method pattern. The intent of this pattern is to
allow the creation of objects without specifying their classes. This is accomplished by
invoking a factory method. The main goal of a factory method is to create an instance of a
class.

You use the factory method pattern for the following scenarios:

When the class is unable to anticipate the type of object that must be instantiated
When the subclass must specify the type of object to instantiate
When the class controls the instantiation of its objects

Consider the following diagram:

[ 185 ]
End-to-End System Testing Chapter 7

As you can see from the preceding diagram, you have the following items:

Factory, which provides the interface for the FactoryMethod() that returns a
type
ConcreteFactory, which overrides or implements the FactoryMethod() to
return a concrete type
ConcreteObject, which inherits or implements the base class or interface

Now is a good time for a demonstration. Imagine that you have three different customers.
Each customer requires using a different relational database as the backend data source.
The databases used by your customers will be Oracle Database, SQL Server, and MySQL.

As a part of your E2E testing, you will need to test against each of these data sources. But
how can you write the program once and have it work against any of those databases? This
is where the Factory method pattern comes in.

Either during the installation process or via the initial configuration of your application,
you can have the user specify the database that they wish to use as the data source. This
information can be stored in a configuration file as an encrypted database connection
string. When your application starts up, it will read the database connection string and
decrypt it. The database connection string will then be passed into the factory method.
Lastly, an appropriate database connection object will be selected, instantiated, and
returned for use by your application.

Now that you have some background, let's create a .NET Framework Console Application
in Visual Studio and call it CH07_Factories. Replace the code in the App.cong file with
the following:
<?xml version="1.0" encoding="utf-8" ?>
<configuration>
<startup>
<supportedRuntime version="v4.0" sku=".NETFramework,Version=v4.8" />
</startup>
<connectionStrings>
<clear />
<add name="SqlServer"
connectionString="Data Source=SqlInstanceName;Initial
Catalog=DbName;Integrated Security=True"
providerName="System.Data.SqlClient"
/>
<add name="Oracle"
connectionString="Data Source=OracleInstance;User
Id=usr;Password=pwd;Integrated Security=no;"
providerName="System.Data.OracleClient"
/>

[ 186 ]
End-to-End System Testing Chapter 7

As you can see, the preceding code has added the connectionStrings element to the
configuration file. Within that element, we clear any existing connection strings and then
add the three database connection strings we will be using for the application. To simplify
the contents of this section, we have unencrypted connection strings, but in the production
environment, make sure that your connection strings are encrypted!

In this project, we will not be using the Main() method in the Program class. We will start
the Factory class, as follows:
namespace CH07_Factories
{
public abstract class Factory
{
public abstract IDatabaseConnection FactoryMethod();
}
}

The preceding code is our abstract factory with a single abstract FactoryMethod() that
returns a type of IDatabaseConnection. Since it does not exist, we'll add that next:
namespace CH07_Factories
{
public interface IDatabaseConnection
{
string ConnectionString { get; }
void OpenConnection();
void CloseConnection();
}
}

In this interface, we have a read-only connection string, a method called

OpenConnection() to open a database connection, and a method called
CloseConnection() to close an open database connection. So far, we have our abstract
Factory and our IDatababaseConnection interface. Next, we will create our concrete
database connection classes. Let's start with the SQL Server database connection class:
public class SqlServerDbConnection : IDatabaseConnection
{
public string ConnectionString { get; }

[ 187 ]
End-to-End System Testing Chapter 7

public SqlServerDbConnection(string connectionString)

{
ConnectionString = connectionString;
}
public void CloseConnection()
{
Console.WriteLine("SQL Server Database Connection Closed.");
}
public void OpenConnection()
{
Console.WriteLine("SQL Server Database Connection Opened.");
}
}

As you can see, the SqlServerDbConnection class fully implements the

IDatabaseConnection interface. The constructor takes connectionString as a single
parameter. The read-only ConnectionString property is then assigned to
connectionString. The OpenConnection() method only prints to the console.

In a real implementation, however, the connection string would be used to connect to the
valid data source specified in the string. Once a database connection is open, it must be
closed. The closing of the database connection would be carried out by the
CloseConnection() method. Next, we repeat the preceding process for the Oracle
database connection and the MySQL database connection:
public class OracleDbConnection : IDatabaseConnection
{
public string ConnectionString { get; }
public OracleDbConnection(string connectionString)
{
ConnectionString = connectionString;
}
public void CloseConnection()
{
Console.WriteLine("Oracle Database Connection Closed.");
}
public void OpenConnection()
{
Console.WriteLine("Oracle Database Connection Closed.");
}
}

[ 188 ]
End-to-End System Testing Chapter 7

We now have the OracleDbConnection class in place. So, the last class we need to
implement is the MySqlDbConnection class:
public class MySqlDbConnection : IDatabaseConnection
{
public string ConnectionString { get; }
public MySqlDbConnection(string connectionString)
{
ConnectionString = connectionString;
}
public void CloseConnection()
{
Console.WriteLine("MySQL Database Connection Closed.");
}
public void OpenConnection()
{
Console.WriteLine("MySQL Database Connection Closed.");
}
}

With that, we have added our concrete classes. The only thing left to do is to create our
ConcreteFactory class that inherits the abstract Factory class. You will need to reference
the System.Configuration.ConfigurationManager NuGet packet:
using System.Configuration;

namespace CH07_Factories
{
public class ConcreteFactory : Factory
{
private static ConnectionStringSettings _connectionStringSettings;

public ConcreteFactory(string connectionStringName)

{
GetDbConnectionSettings(connectionStringName);
}

private static ConnectionStringSettings

GetDbConnectionSettings(string connectionStringName)
{
return
ConfigurationManager.ConnectionStrings[connectionStringName];
}
}
}

[ 189 ]
End-to-End System Testing Chapter 7

As we can see, the class uses the System.Configuration namespace. The

ConnectionStringSettings values are stored in the _connectionStringSettings
member variable. This is set in the constructor that takes connectionStringName. The
name is passed into the GetDbConnectionSettings() method. The quick among you
will see an obvious mistake in the constructor.

The method is getting called but the member variable is not being set. However, we will
pick up this oversight and fix it when we come to run the tests that we have yet to write.
The GetDbConnectionSettings() methods uses ConfigurationManager to read the
required connection string from the ConnectionStrings[] array.

Now, it is time to complete our ConcreteClass by adding FactoryMethod():

public override IDatabaseConnection FactoryMethod()
{
var providerName = _connectionStringSettings.ProviderName;
var connectionString = _connectionStringSettings.ConnectionString;
switch (providerName)
{
case "System.Data.SqlClient":
return new SqlServerDbConnection(connectionString);
case "System.Data.OracleClient":
return new OracleDbConnection(connectionString);
case "System.Data.MySqlClient":
return new MySqlDbConnection(connectionString);
default:
return null;
}
}

Our FactoryMethod() returns a concrete class of type IDatabaseConnection. At the

start of the class, the member variable is read and the values are stored locally for
providerName and connectionString. A switch is then used to determine what type of
database connection to build and pass back.

We are now in a position to test our factory to see whether it works with the different types
of databases used by our customers. This test can be done manually, but for the purpose of
this exercise, we are going to write automation tests.

[ 190 ]
End-to-End System Testing Chapter 7

Create a new NUnit test project. Add a reference to the CH07_Factories project. Then,
add the System.Configuration.ConfigurationManager NuGet package. Rename the
class to UnitTests.cs. Now, add the first test, as shown:
[Test]
public void IsSqlServerDbConnection()
{
var factory = new ConcreteFactory("SqlServer");
var connection = factory.FactoryMethod();
Assert.IsInstanceOf<SqlServerDbConnection>(connection);
}

This test is for a SQL Server database connection. It creates a new

ConcreteFactory() instance and passes in the connectionStringName value of
"SqlServer". The factory then instantiates and returns the correct database connection
object via FactoryMethod(). Finally, the connection object is asserted to test that it is
indeed an instance of type SqlServerDbConnection. We need to write the previous test
twice more for the other database connections, so let's now add the Oracle database
connection test:
[Test]
public void IsOracleDbConnection()
{
var factory = new ConcreteFactory("Oracle");
var connection = factory.FactoryMethod();
Assert.IsInstanceOf<OracleDbConnection>(connection);
}

The test passes in the connectionStringName value of "Oracle". An assertion is made to

test whether the connection object returned is of type OracleDbConnection. Last of all, we
have our MySQL database connection test:
[Test]
public void IsMySqlDbConnection()
{
var factory = new ConcreteFactory("MySQL");
var connection = factory.FactoryMethod();
Assert.IsInstanceOf<MySqlDbConnection>(connection);
}

[ 191 ]
End-to-End System Testing Chapter 7

The test passes in the connectionStringName value of "MySQL". An assertion is made to

test whether the connection object returned is of type MySqlDbConnection. If we run our
tests now, they will all fail because the _connectionStringSettings variable is not
getting set, so let's fix this. Modify your ConcreteFactory constructor as follows:
public ConcreteFactory(string connectionStringName)
{
_connectionStringSettings =
GetDbConnectionSettings(connectionStringName);
}

If you run all your tests now, they should work. If your connection string is not getting
picked up by NUnit, then it will be looking in a different App.config file to what you are
expecting. Add the following line before the line that reads the connection string:
var filepath =
ConfigurationManager.OpenExeConfiguration(ConfigurationUserLevel.None).File
Path;

This will inform you where NUnit is looking for your connection string settings. If the file
does not exist, you can create it manually and duplicate the contents from your main
App.config file. But the problem with this is that the file will more than likely get deleted
upon the next build. So to make the change permanent, you can add a post-build event
command line to your test project.

To do this, right-click on your test project and select Properties. Then on the Properties tab,
select Build Events. In the post-build event command line, add the following command:
xcopy "$(ProjectDir)App.config" "$(ProjectDir)bin\Debug\netcoreapp3.1\" /Y
/I /R

The following screenshot shows the Build Events page of the Project Properties dialog
with the Post-build event command line in place:

[ 192 ]
End-to-End System Testing Chapter 7

This will create the missing file in the test project output folder. The file on your system
may be named testhost.x86.dll.config, since it is on my system. Now, your builds
should be working.

If you change the return type of one of the cases in FactoryMethod(), you will see that
your test fails, as shown in the following screenshot:

[ 193 ]
End-to-End System Testing Chapter 7

Change the code back to the correct type so that your code now passes.

We have seen how to manually E2E test a system, along with how to employ software
factories, and how we can automatically test whether our factories function as expected.
Now we will look at dependency injection and how this can be E2E tested.

Dependency injection
Dependency Injection (DI) helps you to produce code that is loosely coupled by
separating the code's behavior from its dependencies, which leads to more readable code
that is easy to test, extend, and maintain. Code is more readable because you follow the
single responsibility principle. This also leads to much smaller code. Smaller code is easier
to maintain and test, and because we rely upon abstractions instead of on implementations,
we can extend the code more easily according to our needs.

The following are the types of DI that you can implement:

Constructor injection
Property/setter injection
Method injection

Poor man's DI is composed without a container. However, the recommended and best
practice is to use a DI container. In simple terms, a DI container is a registration framework
that instantiates dependencies and injects them when requested.

We are now going to write our own dependency container, interface, services, and client for
our DI example. Then we will write our tests for the dependency project. Bear in mind that
even though tests should be written first, in most business situations I have encountered,
they are written once the software has been written! So in this scenario, we will write our
tests after the software we want has been coded. This can often happen when you employ
multiple teams where some utilize TDD and some don't, or you use third-party code for
which no tests exist.

We mentioned earlier that E2E is best done manually and that automation is hard, but you
can automate tests of the system, as well as performing manual testing. This is particularly
useful if you target multiple data sources.

The first thing you need to have in place is a dependency container. The dependency
container keeps a register of types and instances. You register types before you use them.
When it is time to use an instance of an object, you resolve it into a variable and inject (pass
it) into the constructor, method, or property.

[ 194 ]
End-to-End System Testing Chapter 7

Create a new class library and call it CH07_DependencyInjection. Add a new class called
DependencyContainer, and add the following code:

public static readonly IDictionary<Type, Type> Types = new Dictionary<Type,

Type>();
public static readonly IDictionary<Type, object> Instances = new
Dictionary<Type, object>();

public static void Register<TContract, TImplementation>()

{
Types[typeof(TContract)] = typeof(TImplementation);
}

public static void Register<TContract, TImplementation>(TImplementation

instance)
{
Instances[typeof(TContract)] = instance;
}

In this code, we have two dictionaries that house the types and the instances. We also have
two methods. One is used to register our types, and the second is used to register our
instances. Now that we have the code to register and store our types and instances, we need
a way to resolve them at runtime. Add the following code to the DependencyContainer
class:
public static T Resolve<T>()
{
return (T)Resolve(typeof(T));
}

This method is passed in a type. It calls the method to resolve the type and returns an
instance of that type. So, let's add that method now:
public static object Resolve(Type contract)
{
if (Instances.ContainsKey(contract))
{
return Instances[contract];
}
else
{
Type implementation = Types[contract];
ConstructorInfo constructor = implementation.GetConstructors()[0];
ParameterInfo[] constructorParameters =
constructor.GetParameters();
if (constructorParameters.Length == 0)
{

[ 195 ]
End-to-End System Testing Chapter 7

return Activator.CreateInstance(implementation);
}
List<object> parameters = new
List<object>(constructorParameters.Length);
foreach (ParameterInfo parameterInfo in constructorParameters)
{
parameters.Add(Resolve(parameterInfo.ParameterType));
}
return constructor.Invoke(parameters.ToArray());
}
}

The Resolve() method checks to see whether the Instances dictionary contains an
instance whose key matches the contract. If it does, then that instance is returned.
Otherwise, a new instance is created and returned.

Now, we need an interface that our services to be injected will implement. We'll call it
IService. It will have a single method that will return a string, and the method will be
called WhoAreYou():
public interface IService
{
string WhoAreYou();
}

Our services to be injected will implement the aforementioned interface. Our first class will
be named ServiceOne, and the method will return the string
"CH07_DependencyInjection.ServiceOne()":

public class ServiceOne : IService

{
public string WhoAreYou()
{
return "CH07_DependencyInjection.ServiceOne()";
}
}

The second service is the same except it is called ServiceTwo, and the method returns the
string "CH07_DependencyInjection.ServiceTwo()":
public class ServiceTwo : IService
{
public string WhoAreYou()
{
return "CH07_DependencyInjection.ServiceTwo()";
}
}

[ 196 ]
End-to-End System Testing Chapter 7

The dependency container, interface, and service classes are now in place. Finally, we are
going to add the client that will be used as the demonstration object that will consume our
services via DI. Our class will demonstrate constructor injection, property injection, and
method injection. Add the following code to the top of the class:
private IService _service;

public Client() { }

The _service member variable will be used to store our injected service. We have a
default constructor so that we can test our property and method injection. Add the
constructor that accepts and sets the IService member:
public Client (IService service)
{
_service = service;
}

Next, we will add our property to test property injection and constructor injection:
public IService Service
{
get { return _service; }
set
{
_service = value;
}
}

Then we'll add a method that calls WhoAreYou() on the injected object. The Service
property allows the _service member variable to be set and retrieved. Finally, we will
add our GetServiceName() method:
public string GetServiceName(IService service)
{
return service.WhoAreYou();
}

The GetServiceName() method is called on the injected instance of the IService class.
This method returns the fully qualified name of the service passed in. Now we will write
the unit tests to test the functionality. Add a test project and reference the dependency
project. Call the test project CH07_DependencyInjection.Tests and rename
UnitTest1 to UnitTests.

[ 197 ]
End-to-End System Testing Chapter 7

We will write tests to check that our registration and resolving of instances works, and that
the correct classes are injected by constructor injection, setter injection, and method
injection. Our tests will test the injection of ServiceOne and ServiceTwo. Let's start by
writing our Setup() method as follows:
[TestInitialize]
public void Setup()
{
DependencyContainer.Register<ServiceOne, ServiceOne>();
DependencyContainer.Register<ServiceTwo, ServiceTwo>();
}

In our Setup() method, we register both our implementations of the IService class, these
being ServiceOne() and ServiceTwo(). Now we will write our two test methods to test
the dependency container:
[TestMethod]
public void DependencyContainerTestServiceOne()
{
var serviceOne = DependencyContainer.Resolve<ServiceOne>();
Assert.IsInstanceOfType(serviceOne, typeof(ServiceOne));
}

[TestMethod]
public void DependencyContainerTestServiceTwo()
{
var serviceTwo = DependencyContainer.Resolve<ServiceTwo>();
Assert.IsInstanceOfType(serviceTwo, typeof(ServiceTwo));
}

Both these methods call the Resolve() method. The method checks for an instance of a
type. If an instance exists, it returns it. Otherwise, one is instantiated and returned. It's time
to write the constructor injection tests for serviceOne and serviceTwo:
[TestMethod]
public void ConstructorInjectionTestServiceOne()
{
var serviceOne = DependencyContainer.Resolve<ServiceOne>();
var client = new Client(serviceOne);
Assert.IsInstanceOfType(client.Service, typeof(ServiceOne));
}

[TestMethod]
public void ConstructorInjectionTestServiceTwo()
{
var serviceTwo = DependencyContainer.Resolve<ServiceTwo>();
var client = new Client(serviceTwo);

[ 198 ]
End-to-End System Testing Chapter 7

Assert.IsInstanceOfType(client.Service, typeof(ServiceTwo));
}

In both of these constructor test methods, we resolve the relevant service from the container
registry. Then we pass the service into the constructor. Finally, using the get Service
property, we assert that the service passed in via the constructor is an instance of the
expected service. Let's write the test to show that the property setter injection works as
expected:
[TestMethod]
public void PropertyInjectTestServiceOne()
{
var serviceOne = DependencyContainer.Resolve<ServiceOne>();
var client = new Client();
client.Service = serviceOne;
Assert.IsInstanceOfType(client.Service, typeof(ServiceOne));
}

[TestMethod]
public void PropertyInjectTestServiceTwo()
{
var serviceTwo = DependencyContainer.Resolve<ServiceTwo>();
var client = new Client();
client.Service = serviceTwo;
Assert.IsInstanceOfType(client.Service, typeof(ServiceOne));
}

To test that the setter injection resolves the class we are after, create a client using the
default constructor, then assign the resolved instance to the Service property. Next, we
assert whether the service is an instance of the expected type or not. Finally, for our tests,
we just need to test our method injection:
[TestMethod]
public void MethodInjectionTestServiceOne()
{
var serviceOne = DependencyContainer.Resolve<ServiceOne>();
var client = new Client();
Assert.AreEqual(client.GetServiceName(serviceOne),
"CH07_DependencyInjection.ServiceOne()");
}

[TestMethod]
public void MethodInjectionTestServiceTwo()
{
var serviceTwo = DependencyContainer.Resolve<ServiceTwo>();
var client = new Client();
Assert.AreEqual(client.GetServiceName(serviceTwo),

[ 199 ]
End-to-End System Testing Chapter 7

"CH07_DependencyInjection.ServiceTwo()");
}

Here, we again resolve our instance. Create a new client using the default constructor and
assert passing in the resolved instance and that calling the GetServiceName() method
returns the correct identity of the passed-in instance.

Modularization
A system consists of one or more modules. When a system con two or more modules, you
need to test the interaction between them to make sure they work together as expected.
Let's consider the system for an API shown in the following diagram:

As you can see from the previous diagram, we have a client that accesses a data store in the
cloud via an API. The client sends a request to the HTTP server. The request is
authenticated. Once it has been authenticated, the request is then authorized to access the
API. The data sent by the client is deserialized and then passed on to the business layer. The
business layer then performs either a read, insert, update, or delete operation on the data
store. The data is then passed back to the client from the database via the business layer,
followed by the serialization layer, and then back to the client.

[ 200 ]
End-to-End System Testing Chapter 7

As you can see, we have a number of modules that interact with each other. We have the
following:

Security (Authentication and Authorization) interacting with serialization

(Serialization and Deserialization)
Serialization interacting with the business layer that contains all the business
logic
The Business Logic layer interacting with the data store

If we look at these three preceding points, we can see that a number of tests can be written
to automate the E2E testing process. Many tests are essentially unit tests that become
incorporated into our suite of integration tests. Let's consider some now. We are able to test
the following:

Correct login
Incorrect login
Authorized access
Unauthorized access
Serialization of data
Deserialization of data
Business logic
Database read
Database update
Database insert
Database delete

As you can see from these tests, they are unit tests over integration tests. So, what
integration tests could we write? Well, we could write the following tests:

Send a read request.

Send an insert request.
Send an edit request.
Send a delete request.

Those four tests could be written using the correct username and password and well-
formed data requests, and they could also be written for invalid usernames or passwords
and malformed data requests.

[ 201 ]
End-to-End System Testing Chapter 7

So, you can perform integration testing by using unit tests to test the code in each module,
then using tests that only test the interaction between two modules at a time. You can also
write tests that perform a full E2E operation.

But despite being able to test all this with code, the one thing you must do is run through
the system manually to verify that everything works as expected.

With all these tests completed successfully, you can have the confidence to release your
code to the production environment.

Now that we have covered E2E testing (also known as integration testing), let's take some
time to summarize what we have learned.

Summary
In this chapter, we looked at what E2E testing is. We saw that we can write automated tests,
but we've also come to understand the importance of manually testing the complete
application from an end user perspective.

When we looked at factories, we saw an example of their use when it comes to database
connectivity. We considered a scenario where our app will enable users to use a database of
their choice. We load in a connection string, and then based on that connection string, the
relevant database connection object is instantiated and returned for use. We saw how we
could test our factories for each use case for each different database. Factories can be used
in a number of different scenarios, and now you know what they are, how to use them, and
most importantly, you know how to test them.

DI enables a single class to work with multiple different implementations of an interface.

We saw this in action when we wrote our own dependency container. The interface we
created was implemented by two classes, added to the dependency register, and resolved
when called upon by the dependency container. We implemented unit tests to test the
different implementations for constructor injection, property injection, and method
injection.

Then, we looked at modules. A simple application may consist of a single module, but the
more an application grows in complexity, the more modules will make up that application.
As the number of modules grows, so does the opportunity for something to go wrong.
Therefore, it is very important to test the interaction between modules. The modules
themselves can be tested using unit tests. The interaction between modules can be tested
with more involved tests that run through a complete scenario from start to finish.

[ 202 ]
End-to-End System Testing Chapter 7

In the next chapter, we will be looking at best practices when working with threading and
concurrency. But first, let's test your knowledge on the contents of this chapter.

Questions
1. What is E2E testing?
2. What is another term for E2E testing?
3. What methods should we employ during E2E testing?
4. What are factories, and why do we use them?
5. What is DI?
6. Why should we use a dependency container?

Further reading
The book Dependency Injection in .NET by Manning will introduce you to .NET DI
before guiding you through the various DI frameworks.

[ 203 ]
8
Threading and Concurrency
A process is essentially a program that is executing on an operating system. This process is
made up of more than one thread of execution. A thread of execution is a set of commands
issued by a process. The ability to execute more than one thread at a time is known as
multi-threading. In this chapter, we are going to look at multi-threading and concurrency.

Multiple threads are allotted a set amount of time to execute, and each thread is executed
on a rotational basis by a thread scheduler. The thread scheduler schedules the threads
using a technique called time slicing and then passes each thread to the CPU to be executed
at the scheduled time.

Concurrency is the ability to run more than one thread at exactly the same time. This can be
accomplished on computers with more than one processor core. The more processor cores a
computer has, the more threads of execution can be executed concurrently.

As we look at concurrency and threading in this chapter, we will encounter the problems of
blocking, deadlocks, and race conditions. You will see how we can overcome these
problems using clean coding techniques.

In the course of this chapter, we will cover each of the following topics:

Understanding the thread life cycle

Adding thread parameters
Using a thread pool
Using a mutual exclusion object with synchronous threads
Working with parallel threads using semaphores
Limiting the number of processors and threads in the thread pool
Preventing deadlocks
Preventing race conditions
Understanding static constructors and methods
Mutability, immutability, and thread safety
Threading and Concurrency Chapter 8

Synchronized method dependencies

Using the Interlocked class for simple state changes
General recommendations

After working through this chapter and developing your threading and concurrency skills,
you will have acquired the following skills:

The ability to understand and discuss the thread life cycle

An understanding of and ability to use foreground and background threads
The ability to throttle threads and set the number of processors to use
concurrently using a thread pool
The ability to understand the effects of static constructors and methods in
relation to multi-threading and concurrency
The ability to take into account mutability and immutability and their impact on
thread safety
The ability to understand what causes race conditions and how to avoid them
The ability to understand what causes deadlocks and how to avoid them
The ability to perform simple state changes using the Interlocked class

To run through the code in this chapter, you will need a .NET Framework console
application. Unless otherwise stated, all code will be placed in the Program class.

Understanding the thread life cycle

Threads in C# have an associated life cycle. The life cycle for threads is as follows:

[ 205 ]
Threading and Concurrency Chapter 8

When a thread starts, it enters the running state. When running, the thread can enter
a wait, sleep, join, stop, or suspended state. Threads can also be aborted. Aborted threads
enter the stop state. You can suspend and resume a thread by calling the Suspend() and
Resume() methods, respectively.

A thread will enter the wait state when the Monitor.Wait(object obj) method is called.
The thread will then continue when the Monitor.Pulse(object obj) method is called.
Threads enter sleep mode by calling the Thread.Sleep(int millisecondsTimeout)
method. Once the elapsed time has passed, the thread returns to the running state.

The Thread.Join() method causes a thread to enter the wait state. A joined thread will
remain in the wait state until all dependent threads have finished running, upon which it
will enter the running state. However, if any dependent threads are aborted, then this
thread is also aborted and enters the stop state.

Threads that have completed or have been aborted cannot be restarted.

Threads can run in the foreground or the background. Let's look at both foreground and
background threads, starting with foreground threads:

Foreground threads: By default, threads run in the foreground. A process will

continue to run while at least one foreground thread is currently running. Even
if Main() completes but a foreground thread is running, the application process
will remain active until the foreground thread terminates. Creating a foreground
thread is really simple, as the following code shows:
var foregroundThread = new Thread(SomeMethodName);
foregroundThread.Start();

Background threads: You create a background thread in the same way that you
create foreground threads, except that you also have to explicitly set a thread to
run in the background, as shown:
var backgroundThread = new Thread(SomeMethodName);
backgroundThread.IsBackground = true;
backgroundThread.Start();

[ 206 ]
Threading and Concurrency Chapter 8

Background threads are used to carry out background tasks and keep the user interface
responsive to the user. When the main process terminates, any background threads that are
executing are also terminated. However, even if the main process terminates, any
foreground threads that are running will run to completion.

In the next section, we will look at thread parameters.

Adding thread parameters

Methods that run in threads often have parameters. So, when executing a method within a
thread, it is useful to know how to pass the method parameters into the thread.

Let's say that we have the following method, which adds two integers together and returns
a result:
private static int Add(int a, int b)
{
return a + b;
}

As you can see, the method is simple. There are two parameters called a and b. These two
parameters will need to be passed into the thread for the Add() method to run properly.
We will add an example method that will do just that:
private static void ThreadParametersExample()
{
int result = 0;
Thread thread = new Thread(() => { result = Add(1, 2); });
thread.Start();
thread.Join();
Message($"The addition of 1 plus 2 is {result}.");
}

In this method, we declare an integer with an initial value of 0. We then create a new thread
that calls the Add() method with the 1 and 2 parameter values, and then assign the result
to the integer variable. The thread then starts and we wait for it to finish executing by
calling the Join() method. Finally, we print the result to the console window.

Let's add our Message() method:

internal static void Message(string message)
{
Console.WriteLine(message);
}

[ 207 ]
Threading and Concurrency Chapter 8

The Message() method simply takes a string and outputs it to the console window. All we
have to do now is update the Main() method, as follows:
static void Main(string[] args)
{
ThreadParametersExample();
Message("=== Press any Key to exit ===");
Console.ReadKey();
}

In our Main() method, we call our example method and then wait for the user to press any
key before exiting. You should see the following output:

As you can see, 1 and 2 were the method parameters passed into the addition method, and
3 was the value returned by the thread. The next topic we will look at is using a thread
pool.

Using a thread pool

A thread pool improves performance by creating a collection of threads during application
initialization. When a thread is required, it is assigned a single task. That task will be
executed. Once executed, the thread is returned to the thread pool to be reused.

Since thread creation is expensive in .NET, we can improve performance by using a thread
pool. Each process has a fixed number of threads based on the system resources available,
such as memory and the CPU. However, we can increase or decrease the number of threads
used by the thread pool. It is normally best to let the thread pool take care of how many
threads to use, rather than manually setting these values.

The different ways to create a thread pool are as follows:

Using the Task Parallel Library (TPL) (on .NET Framework 4.0 and higher)
Using ThreadPool.QueueUserWorkItem()
Using asynchronous delegates
Using BackgroundWorker

[ 208 ]
Threading and Concurrency Chapter 8

As a rule of thumb, you should only use a thread pool for server-side
applications. For client-side applications, use foreground and background
threads, as necessary.

In this book, we will just look at the TPL and the QueueUserWorkItem() method. You can
check out how to use the other two methods at http://www.albahari.com/threading/.
We'll look at the TPL next.

Task Parallel Library

An asynchronous operation in C# is represented by a task. A task in C# is represented by
the Task class in the TPL. As you will gather from the name, task parallelism enables
multiple tasks to be executed concurrently, which we will learn about in the following
subsections. The first Parallel class method we will look at is the Invoke() method.

Parallel.Invoke()
In our first example, we will invoke three separate methods using Parallel.Invoke().
Add the following three methods:
private static void MethodOne()
{
Message($"MethodOne Executed: Thread
Id({Thread.CurrentThread.ManagedThreadId})");
}

private static void MethodTwo()

{
Message($"MethodTwo Executed: Thread
Id({Thread.CurrentThread.ManagedThreadId})");
}

private static void MethodThree()

{
Message($"MethodThree Executed: Thread
Id({Thread.CurrentThread.ManagedThreadId})");
}

[ 209 ]
Threading and Concurrency Chapter 8

As you can see, these three methods are almost identical, apart from their names and the
message printed to the console window via the Message() method we wrote earlier. Now,
we'll add the UsingTaskParallelLibrary() method to execute these three methods in
parallel:
private static void UsingTaskParallelLibrary()
{
Message($"UsingTaskParallelLibrary Started: Thread Id =
({Thread.CurrentThread.ManagedThreadId})");
Parallel.Invoke(MethodOne, MethodTwo, MethodThree);
Message("UsingTaskParallelLibrary Completed.");
}

In this method, we write a message to the console window indicating the start of the
method. We then invoke the MethodOne, MethodTwo, and MethodThree methods in
parallel. Then, we write a message to the console window indicating that the method has
reached its end, and then we wait for a key to be pressed before exiting the method. Run
the code and you should see the following output:

In the preceding screenshot, you can see that thread one is reused. Let's now move on to the
Parallel.For() loop.

Parallel.For()
In our next TPL example, we will look at a simple Parallel.For() loop. Add the
following method to the Program class of a new .NET Framework console application:
private static void Method()
{
Message($"Method Executed: Thread
Id({Thread.CurrentThread.ManagedThreadId})");
}

[ 210 ]
Threading and Concurrency Chapter 8

All this method does is output a string to the console window. We'll now create the method
that executes the Parallel.For() loop:
private static void UsingTaskParallelLibraryFor()
{
Message($"UsingTaskParallelLibraryFor Started: Thread Id =
({Thread.CurrentThread.ManagedThreadId})");
Parallel.For(0, 1000, X => Method());
Message("UsingTaskParallelLibraryFor Completed.");
}

In this method, we loop through 0 to 1000, calling Method(). You will see how the threads
are reused with the different method calls, as in the following screenshot:

Now, we will look at using the ThreadPool.QueueUserWorkItem() method.

[ 211 ]
Threading and Concurrency Chapter 8

ThreadPool.QueueUserWorkItem()
The ThreadPool.QueueUserWorkItem() method accepts a WaitCallback method and
queues it ready for execution. WaitCallback is a delegate that represents a callback
method to be executed by a thread pool thread. When a thread becomes available, the
method is executed. Let's add a simple example. We'll start by adding
WaitCallbackMethod:

private static void WaitCallbackMethod(Object _)

{
Message("Hello from WaitCallBackMethod!");
}

This method accepts a type of object. However, since the parameter will be unused, we use
the discard variable (_). A message is printed to the console window. Now, all we need is
the code to call the method:
private static void ThreadPoolQueueUserWorkItem()
{
ThreadPool.QueueUserWorkItem(WaitCallbackMethod);
Message("Main thread does some work, then sleeps.");
Thread.Sleep(1000);
Message("Main thread exits.");
}

As you can see, we use the ThreadPool class to queue WaitCallbackMethod() in the
thread pool via the call to the QueueUserWorkItem() method. We then do some work on
the main thread. The main thread then goes to sleep. A thread becomes available from the
thread pool and WaitCallBackMethod() is executed. The thread is then returned back to
the thread pool to be reused. Execution returns to the main thread, which then completes
and terminates.

In the next section, we will discuss thread-locking objects, known as Mutual Exclusion
Objects (mutexes).

Using a mutex with synchronous threads

In C#, a mutex is a thread-locking object that works across multiple processes. Only a
process that can request or release a resource can modify the mutex. When a mutex is
locked, the process will have to wait in a queue. When the mutex is unlocked, then it can be
accessed. Multiple threads can use the same mutex, but only in a synchronous manner.

[ 212 ]
Threading and Concurrency Chapter 8

The benefits of using a mutex are that a mutex is a simple lock obtained before entering a
critical piece of code. That lock is released when the critical piece of code is exited. Because
only a single thread is in the critical piece of code at any one time, the data will remain in a
consistent state as there will be no race conditions.

There are several disadvantages to using a mutex:

Thread starvation occurs when a thread is unable to move forward as an existing

thread has obtained a lock and has either gone to sleep or is pre-empted
(prevented from completing its task).
When a mutex is locked, only the thread that obtained the lock can unlock it. No
other thread can lock or unlock it.
Only one thread at a time is allowed to enter the critical piece of code. CPU time
can be wasted as the normal implementation of a mutex may lead to a busy
waiting state.

We will now write a program that demonstrates the use of a mutex. Start a new .NET
Framework console application. Add the following line to the top of the class:
private static readonly Mutex _mutex = new Mutex();

Here, we have declared a primitive called _mutex, which we will use for inter-process
synchronization. Now, add a method to demonstrate thread synchronization using a
mutex:
private static void ThreadSynchronisationUsingMutex()
{
try
{
_mutex.WaitOne();
Message($"Domain Entered By: {Thread.CurrentThread.Name}");
Thread.Sleep(500);
Message($"Domain Left By: {Thread.CurrentThread.Name}");
}
finally
{
_mutex.ReleaseMutex();
}
}

[ 213 ]
Threading and Concurrency Chapter 8

In this method, the current thread is blocked until the current wait handle receives a signal.
Then, when the signal is given, it is safe for the next thread to enter. Upon completion,
other threads are unblocked from trying to gain ownership of the mutex. Next, add the
MutexExample() method:

private static void MutexExample()

{
for (var i = 1; i <= 10; i++)
{
var thread = new Thread(ThreadSynchronisationUsingMutex)
{
Name = $"Mutex Example Thread: {i}"
};
thread.Start();
}
}

In this method, we create 10 threads and start them. Each thread executes the
ThreadSynchronisationUsingMutex() method. Now, finally, update the Main()
method:
static void Main(string[] args)
{
SemaphoreExample();
Console.ReadKey();
}

The Main() method runs our mutex example. The output should be similar to the one in
the following screenshot:

[ 214 ]
Threading and Concurrency Chapter 8

Run the example again and you may end up with different thread numbers. If they are the
same numbers, then they may be in different orders.

Now that we have looked at mutexes, let's look at semaphores.

Working with parallel threads using

semaphores
In multi-threaded applications, a non-negative number, known as a semaphore, is shared
between threads that have a number of 1 or 2. In terms of synchronization, 1
specifies wait and 2 specifies signal. We can associate a semaphore with a number of buffers,
which can each be worked on simultaneously by different processes.

So, essentially, semaphores are signaling mechanisms of the integer and binary primitive
types that can be modified by wait and signal operations. If there are no free resources, then
processes that require a resource should execute the wait operation until the semaphore
value is greater than 0. Semaphores can have multiple program threads and they can be
changed by any object, obtaining a resource or releasing it.

The advantages of using semaphores are down to the fact that more than one thread can
access the critical piece of code. A semaphore is executed in the kernel and is machine-
independent. The critical piece of code is protected from multiple processes if you use
semaphores. Unlike a mutex, a semaphore never wastes processing time and resources.

Just like a mutex, semaphores also have their own set of disadvantages. Priority inversion is
one of the biggest disadvantages and occurs when a high-priority thread is forced to wait
for a semaphore to be released by its low-priority owning thread.

This can be further compounded if the low-priority thread is prevented from completing by
mid-priority threads prior to their release. This is known as unbounded priority inversion
because we can no longer predict the delay to the high-priority thread. With semaphores,
the operating system must keep track of all wait and signal calls.

Semaphores are used by convention, but they are not forced. You need to execute wait and
signal operations in the correct order; otherwise, you risk deadlocks in your code. Because
of the complexity of using semaphores, there may be times when a mutual exclusion cannot
be obtained. Loss of modularity in large systems is also another drawback and semaphores
are prone to programming errors that result in deadlocks and mutual exclusion violation.

[ 215 ]
Threading and Concurrency Chapter 8

We're going to write a program now that demonstrates the use of semaphores:
private static readonly Semaphore _semaphore = new Semaphore(2, 4);

We have added a new semaphore variable. The first parameter states the initial number of
requests for the semaphore that can be granted concurrently. The second parameter states
the maximum number of requests for the semaphore that can be granted concurrently. Add
the StartSemaphore() method:
private static void StartSemaphore(object id)
{
Console.WriteLine($"Object {id} wants semaphore access.");
try
{
_semaphore.WaitOne();
Console.WriteLine($"Object {id} gained semaphore access.");
Thread.Sleep(1000);
Console.WriteLine($"Object {id} has exited semaphore.");
}
finally
{
_semaphore.Release();
}
}

The current thread is blocked until the current wait handle receives a signal. The thread can
then do its work. Finally, the semaphore is released and the count returns to the previous
count. Now, add the SemaphoreExample() method:
private static void SemaphoreExample()
{
for (int i = 1; i <= 10; i++)
{
Thread t = new Thread(StartSemaphore);
t.Start(i);
}
}

This example generates 10 threads, which execute the StartSemaphore() method. Let's
update the Main() method to run the code:
static void Main(string[] args)
{
SemaphoreExample();
Console.ReadKey();
}

[ 216 ]
Threading and Concurrency Chapter 8

The Main() method calls SemaphoreExample() and then waits for a user keypress to exit.
You should see the following output:

Let's move on to look at how we limit the number of processors and threads in the thread
pool.

Limiting the number of processors and

threads in the thread pool
There may be times when you need to limit the number of processors and threads used by
your computer program.

[ 217 ]
Threading and Concurrency Chapter 8

To reduce the number of processors that your program uses, you obtain the current process
and set its processor affinity value. For example, say that we have a four-core computer and
we want to limit our usage to the first two cores. The binary value for the first two cores is
11, which is 3 in integer form. Now, let's add a method to a new .NET Framework console
application and call it AssignCores():
private static void AssignCores(int cores)
{
Process.GetCurrentProcess().ProcessorAffinity = new IntPtr(cores);
}

We pass in an integer to the method. This integer value will be converted into a binary
value by .NET Framework. That binary value will use the processors identified by the value
of 1. For binary values of 0, the processors will not be used. So, since machine code is
represented by binary numbers, 0110 (6) will use cores 2 and 3, 1100 (3) will use cores 1
and 2, and 0011 (12) will use cores 3 and 4.

If you want a refresher on binary, refer to https://www.computerhope.

com/jargon/b/binary.htm.

Now, to set the maximum number of threads, we call the SetMaxThreads() method on
the ThreadPool class. This method takes two parameters, which are both integers. The first
parameter is the maximum number of worker threads in the thread pool and the second
parameter is the maximum number of asynchronous I/O threads in the thread pool. We'll
now add our method to set the maximum number of threads:
private static void SetMaxThreads(int workerThreads, int asyncIoThreads)
{
ThreadPool.SetMaxThreads(workerThreads, asyncIoThreads);
}

As you can see, it is pretty straightforward to set thread maximums and

processors in your programs. Most of the time, you will not have to do
this in your programs. The main reason for manually setting the number
of threads and/or processors to use in your program is for if your
programs run into performance issues. If your program does not
experience performance issues, then it is best not to set the number of
threads or the number of processors.

The next topic we will look at is deadlocks.

[ 218 ]
Threading and Concurrency Chapter 8

Preventing deadlocks
A deadlock occurs when two or more threads are executed and are waiting for each other
to finish. This problem manifests in computer programs when they hang. For the end user,
this can be very bad and can result in the loss or corruption of data. An example of this is
executing two batches of data input that crash halfway through a transaction and cannot be
rolled back. This is not good; let me explain why with an example.

Consider a major banking transaction that will take £1 million out of a customer's business
bank account to pay their Her Majesty's Revenue and Customs (HMRC) tax bill. The
money is taken from the business account, but before the money is deposited in the HMRC
bank account, a deadlock occurs. There is no recovery option and so the application has to
be terminated and restarted. As a result, the business bank account is reduced by £1 million
but the HMRC tax bill has not been paid. The customer is still liable to pay the tax bill. But
what happens to the money that has been taken out of the account? So, you can see the
importance of removing the possibility of deadlocks occurring due to the problems they
can cause.

To keep things simple, we will deal with two threads, shown in the following diagram:

We will call our threads Thread 1 and Thread 2 and our resources Resource
1 and Resource 2. Thread 1 obtains a lock on Resource 1. Thread 2 obtains a lock
on Resource 2. Thread 1 requires access to Resource 2 but has to wait because Thread 2 has
locked Resource 2. Thread 2 requires access to Resource 1 but has to wait because Thread
1 has locked Resource 1. This results in both Thread 1 and Thread 2 being in a wait state.
Since neither thread can continue until the other thread releases its resource, both threads
are in a deadlock situation. When a computer program is in a deadlock situation, it hangs,
forcing you to terminate the program.

A code example of a deadlock will be a nice way to illustrate this, and so in the next section,
we will code a deadlock example.

[ 219 ]
Threading and Concurrency Chapter 8

Coding a deadlock example

The best way to understand this is with a working example. We are going to write some
code consisting of two methods that have two different locks each. They will both lock
objects that the other method needs. Because each thread locks the resources that the other
thread needs, they will both enter a deadlock state. Once we have our example working, we
will then modify it so that our code recovers from the deadlock situation and is able to
continue.

Create a new .NET Framework console application and call it CH08_Deadlocks. We will
need two objects as member variables, so let's add them:
static object _object1 = new object();
static object _object2 = new object();

These objects will be used as our lock objects. We will have two threads, and each thread
will execute its own method. Now, add Thread1Method() to your code:
private static void Thread1Method()
{
Console.WriteLine("Thread1Method: Thread1Method Entered.");
lock (_object1)
{
Console.WriteLine("Thread1Method: Entered _object1 lock.
Sleeping...");
Thread.Sleep(1000);
Console.WriteLine("Thread1Method: Woke from sleep");
lock (_object2)
{
Console.WriteLine("Thread1Method: Entered _object2 lock.");
}
Console.WriteLine("Thread1Method: Exited _object2 lock.");
}
Console.WriteLine("Thread1Method: Exited _object1 lock.");
}

Thread1Method() obtains a lock on _object1. It then sleeps for 1 second. When it

awakes, a lock is obtained on _object2. The method then exits both locks and terminates.

[ 220 ]
Threading and Concurrency Chapter 8

Thread2Method() obtains a lock on _object2. It then sleeps for 1 second. When it

awakes, a lock is obtained on _object1. The method then exits both locks and terminates:
private static void Thread2Method()
{
Console.WriteLine("Thread2Method: Thread1Method Entered.");
lock (_object2)
{
Console.WriteLine("Thread2Method: Entered _object2 lock.
Sleeping...");
Thread.Sleep(1000);
Console.WriteLine("Thread2Method: Woke from sleep.");
lock (_object1)
{
Console.WriteLine("Thread2Method: Entered _object1 lock.");
}
Console.WriteLine("Thread2Method: Exited _object1 lock.");
}
Console.WriteLine("Thread2Method: Exited _object2 lock.");
}

Well, we now have our two methods in place to demonstrate a deadlock. We just need the
code to call them in a way that will cause a deadlock. Let's add
the DeadlockNoRecovery() method:
private static void DeadlockNoRecovery()
{
Thread thread1 = new Thread((ThreadStart)Thread1Method);
Thread thread2 = new Thread((ThreadStart)Thread2Method);

thread1.Start();
thread2.Start();

Console.WriteLine("Press any key to exit.");

Console.ReadKey();
}

In the DeadlockNoRecovery() method, we create two threads. Each thread is assigned a

different method. Then, each thread is started. The program is then paused until the user
presses a key. Now, update the Main() method and run your code:
static void Main()
{
DeadlockNoRecovery();
}

[ 221 ]
Threading and Concurrency Chapter 8

When you run your program, you should see the following output:

As you can see, because thread1 has locked _object1, thread2 is blocked from
obtaining a lock on _object1. Also, because thread2 has locked _object2, thread1 is
blocked from obtaining a lock on _object2. So, both threads are in deadlock and the
program hangs.

We will now write some code that demonstrates how to avoid this deadlock situation from
occurring. We will be using the Monitor.TryLock() method to try and obtain a lock
within a certain number of milliseconds. We will then exit a successful lock
with Monitor.Exit().

Now, add the DeadlockWithRecovery() method:

private static void DeadlockWithRecovery()
{
Thread thread4 = new Thread((ThreadStart)Thread4Method);
Thread thread5 = new Thread((ThreadStart)Thread5Method);

thread4.Start();
thread5.Start();

Console.WriteLine("Press any key to exit.");

Console.ReadKey();
}

The DeadlockWithRecovery() method creates two foreground threads. It then starts the
threads, prints a message to the console, and waits for the user to press a key before exiting.
We will now add the code for Thread4Method():
private static void Thread4Method()
{
Console.WriteLine("Thread4Method: Entered _object1 lock.
Sleeping...");
Thread.Sleep(1000);
Console.WriteLine("Thread4Method: Woke from sleep");

[ 222 ]
Threading and Concurrency Chapter 8

if (!Monitor.TryEnter(_object1))
{
Console.WriteLine("Thead4Method: Failed to lock _object1.");
return;
}
try
{
if (!Monitor.TryEnter(_object2))
{
Console.WriteLine("Thread4Method: Failed to lock _object2.");
return;
}
try
{
Console.WriteLine("Thread4Method: Doing work with _object2.");
}
finally
{
Monitor.Exit(_object2);
Console.WriteLine("Thread4Method: Released _object2 lock.");
}
}
finally
{
Monitor.Exit(_object1);
Console.WriteLine("Thread4Method: Released _object2 lock.");
}
}

Thread4Method() sleeps for 1 second. It then tries to get a lock on _object1. If it fails to
get a lock on _object1, it returns from the method. If a lock on _object1 is obtained, then
it tries to get a lock on _object2. If a lock on _object2 can't be obtained, then it returns
from the method. If a lock is obtained on _object2, then it performs the necessary work
on _object2. The lock on _object2 is then released, and then the lock on _object1 is
released.

Our Thread5Method() method does exactly the same thing, except the objects—_object1
and _object2—are locked in reverse order:
private static void Thread5Method()
{
Console.WriteLine("Thread5Method: Entered _object2 lock.
Sleeping...");
Thread.Sleep(1000);
Console.WriteLine("Thread5Method: Woke from sleep");
if (!Monitor.TryEnter(_object2))
{

[ 223 ]
Threading and Concurrency Chapter 8

Console.WriteLine("Thead5Method: Failed to lock _object2.");

return;
}
try
{
if (!Monitor.TryEnter(_object1))
{
Console.WriteLine("Thread5Method: Failed to lock _object1.");
return;
}
try
{
Console.WriteLine("Thread5Method: Doing work with _object1.");
}
finally
{
Monitor.Exit(_object1);
Console.WriteLine("Thread5Method: Released _object1 lock.");
}
}
finally
{
Monitor.Exit(_object2);
Console.WriteLine("Thread5Method: Released _object2 lock.");
}
}

Now, add the DeadlockWithRecovery() method call to your Main() method:

static void Main()
{
DeadlockWithRecovery();
}

Then, run your code a few times. The majority of the time, you will see what is in the
following screenshot, where all the locks have been successfully obtained:

[ 224 ]
Threading and Concurrency Chapter 8

Then, press any key and the program will exit. If you keep running the program, you will
eventually find a lock that fails. The program failed to get a lock
on _object2 in Thread5Method(). However, if you press any key, the program will exit.
As you can see, by using Monitor.TryEnter(), you can try and lock an object. But if the
lock is not obtained, then you are able to take another action without your program
hanging.

In the next section, we look at preventing race conditions.

Preventing race conditions

When multiple threads using the same resource produce different outcomes due to the
timings of each thread, this is known as a race condition. We will demonstrate this in
action now.

In our demonstration, we will have two threads. Each thread will call a method to print the
alphabet. One method will print the alphabet using uppercase letters. The second method
will print the alphabet using lowercase letters. From the demonstration, we'll see how the
output is wrong, and every time the program is run, the output will be wrong.

First, add the ThreadingRaceCondition() method:

static void ThreadingRaceCondition()
{
Thread T1 = new Thread(Method1);
T1.Start();
Thread T2 = new Thread(Method2);
T2.Start();
}

ThreadingRaceCondition() produces two threads and starts them. It also references two
methods. Method1() prints out the alphabet in uppercase and Method2() prints out the
alphabet in lowercase. Let's add Method1() and Method2():
static void Method1()
{
for (_alphabetCharacter = 'A'; _alphabetCharacter <= 'Z';
_alphabetCharacter ++)
{
Console.Write(_alphabetCharacter + " ");
}
}

private static void Method2()

[ 225 ]
Threading and Concurrency Chapter 8

{
for (_alphabetCharacter = 'a'; _alphabetCharacter <= 'z';
_alphabetCharacter++)
{
Console.Write(_alphabetCharacter + " ");
}
}

Both Method1() and Method2() reference the _alphabetCharacter variable. So, add the
member to the top of the class:
private static char _alphabetCharacter;

Now, update MainMethod():

static void Main(string[] args)
{
Console.WriteLine("\n\nRace Condition:");
ThreadingRaceCondition();
Console.WriteLine("\n\nPress any key to exit.");
Console.ReadKey();
}

We now have our code in place to demonstrate the race condition. If you run the program
multiple times, you will see that the results are not what we expect. You should even see
characters that are not part of the alphabet:

Not exactly what we were expecting, is it?

We are going to solve this problem by using the TPL. The aim of the TPL is to simplify
parallelism and concurrency. As most computers today have two or more processors, the
TPL will scale the degree of concurrency dynamically to make the most efficient use of all
the available processors.

[ 226 ]
Threading and Concurrency Chapter 8

The partitioning of work, the scheduling of threads in the thread pool,

cancellation support, state management, and so on are also carried out by
the TPL. A link to the official Microsoft TPL documentation can be found
in the Further reading section of this chapter.

You will see just how simple the solution to the aforementioned problem can be. We have a
task that runs Method1(). The task then continues with Method2(). We then call
Wait() to wait for the task to complete execution. Now, add the
ThreadingRaceConditionFixed() method to your source code:

static void ThreadingRaceConditionFixed()

{
Task
.Run(() => Method1())
.ContinueWith(task => Method2())
.Wait();
}

Modify your Main() method, as follows:

static void Main(string[] args)
{
//Console.WriteLine("\n\nRace Condition:");
//ThreadingRaceCondition();
Console.WriteLine("\n\nRace Condition Fixed:");
ThreadingRaceConditionFixed();
Console.WriteLine("\n\nPress any key to exit.");
Console.ReadKey();
}

Run the code now. If you run it multiple times, you will see that the output is always the
same, as in the following screenshot:

So far, we have seen what a thread is and how to use them in the foreground and
background. We have also looked at deadlocks and how to solve them
with Monitor.TryEnter(). Finally, we looked at what race conditions are and how to
solve them using the TPL.

Now, we will move on to looking at static constructors and methods.

[ 227 ]
Threading and Concurrency Chapter 8

Understanding static constructors and

methods
If multiple classes require access to a property instance simultaneously, then one of the
threads will be requested to run the static constructor (also known as the type initializer).
While waiting for the type initializer to run, all the other threads will be locked. Once the
type initializer has run, the locked threads are unlocked and are able to access
the Instance property.

Static constructors are thread-safe as they are guaranteed to run only once per application
domain. They are executed before accessing any static members and before any class
instantiation is performed.

Should an exception be raised in and escape from a static constructor, then

TypeInitializationException is generated, which causes the CLR to
exit your program.

Before any threads can access a class, static initializers and static constructors must finish
executing.

Static methods only keep a single copy of the method and its data at the type level. This
means that the same method and its data will be shared between different instances. Each
thread in an application has its own stack. Value types passed into static methods are
created on the calling thread's stack, and so they are thread-safe. This means that if two
threads call the same code and pass the same value in, there will be two copies of that
value—one on each thread's stack. So, multiple threads will not affect each other.

However, if you have a static method that accesses a member variable, then it is not thread-
safe. Two different threads call the same method and so both will have access to the
member variable. A process or context-switching occurs between threads; each thread will
access and modify the member variable. This leads to race conditions, as you saw earlier in
this chapter.

You also run into problems if you pass reference types into a static method, as different
threads will have access to the same reference type. This also causes a race condition.

[ 228 ]
Threading and Concurrency Chapter 8

When working with static methods that will be used across threads, avoid
member variable access and do not pass reference types in. Static methods
are thread-safe as long as you pass in primitive types and don't modify
the state.

Now that we've discussed static constructors and methods, we will run through some
example code.

Adding static constructors to our sample code

Start a new .NET Framework console application. Add a class called
StaticConstructorTestClass to the project. Then, add a read-only static string variable
called _message:
public class StaticConstructorTestClass
{
private readonly static string _message;
}

The _message variable is returned to the caller by the Message() method. Let's write the
Message() method now:

public static string Message()

{
return $"Message: {_message}";
}

This method returns the message stored in the _message variable. Now, we need to write
our constructor:
static StaticConstructorTestClass()
{
Console.WriteLine("StaticConstructorTestClass static constructor
started.");
_message = "Hello, World!";
Thread.Sleep(1000);
_message = "Goodbye, World!";
Console.WriteLine("StaticConstructorTestClass static constructor
finished.");
}

[ 229 ]
Threading and Concurrency Chapter 8

In our constructor, we write a message to the screen. We then set the member variable and
let the thread sleep for a second. Then, we set the message again and write another message
to the console. Now, in the Program class, update the Main() method, as follows:
static void Main(string[] args)
{
var program = new Program();
program.StaticConstructorExample();
Thread.CurrentThread.Join();
}

Our Main() method instantiates the Program class. The StaticConstructorExample()

method is then called. When the program halts and we can see the result, we join threads.
You can see the output in the following screenshot:

We'll now take a look at examples of static methods.

Adding static methods to our sample code

We are now going to look at thread-safe static methods and non-thread-safe methods in
action. Add a new class called StaticExampleClass to a new .NET Framework console
application. Then, add the following code:
public static class StaticExampleClass
{
private static int _x = 1;
private static int _y = 2;
private static int _z = 3;
}

[ 230 ]
Threading and Concurrency Chapter 8

At the top of our class, we add three integers—_x, _y, and _z—with values of 1, 2, and 3,
respectively. These variables can be modified between threads. Now, we will add a static
constructor to print out the values of these variables:
static StaticExampleClass()
{
Console.WriteLine($"Constructor: _x={_x}, _y={_y}, _z={_z}");
}

As you can see, the static constructor simply prints out the values of the variables to the
console window. Our first method will be a thread-safe method called
ThreadSafeMethod():

internal static void ThreadSafeMethod(int x, int y, int z)

{
Console.WriteLine($"ThreadSafeMethod: x={x}, y={y}, z={z}");
Console.WriteLine($"ThreadSafeMethod: {x}+{y}+{z}={x+y+z}");
}

This method is thread-safe because it only operates on by value parameters. It does not
interact with the member variables and does not include any by reference values. So, no
matter what values are passed in, you will always get the expected result.

This means that regardless of whether only a single thread or even millions of threads are
accessing the method, the output for each thread will be what you expect when you pass in
the input values, even despite context switching. The following screenshot shows the
output:

[ 231 ]
Threading and Concurrency Chapter 8

Now that we have looked at thread-safe methods, it is only right that we look at non-
thread-safe methods. By now, you know that a static method that operates on by reference
values or static member variables is not thread-safe.

In our next example, we will use a method with the same three parameters as
ThreadSafeMethod(), but this time, we will set the member variables, output a message,
go to sleep for a while, and then awake to print the values out again. Add the following
NotThreadSafeMethod() method to StaticExampleClass:

internal static void NotThreadSafeMethod(int x, int y, int z)

{
_x = x;
_y = y;
_z = z;
Console.WriteLine(
$"{Thread.CurrentThread.ManagedThreadId}-NotThreadSafeMethod:
_x={_x}, _y={_y}, _z={_z}"
);
Thread.Sleep(300);
Console.WriteLine(
$"{Thread.CurrentThread.ManagedThreadId}-ThreadSafeMethod:
{_x}+{_y}+{_z}={_x + _y + _z}"
);
}

In this method, we set the member variables to the values passed into the method. We then
output those values to the console windows and go to sleep for 300 milliseconds. Then,
upon waking from our sleep, we print the values out again. In the Program class, update
the Main() method, as shown:
static void Main(string[] args)
{
var program = new Program();
program.ThreadUnsafeMethodCall();
Console.ReadKey();
}

[ 232 ]
Threading and Concurrency Chapter 8

In the Main() method, we instantiate the program class, call

ThreadUnsafeMethodCall(), and then wait for the user to press a key before exiting. So,
let's add ThreadUnsafeMethodCall() to the Program class:
private void ThreadUnsafeMethodCall()
{
for (var index = 0; index < 10; index++)
{
var thread = new Thread(() =>
{
StaticExampleClass.NotThreadSafeMethod(index + 1, index + 2,
index + 3);
});
thread.Start();
}
}

This method produces 10 threads that call NotThreadSafeMethod() of

StaticExampleClass. If you run the code, you will see an output similar to that in the
following screenshot:

As you can see, the output is not what we would expect. This is because of the pollution
from different threads. This leads us nicely to the next section on mutability, immutability,
and thread safety.

[ 233 ]
Threading and Concurrency Chapter 8

Mutability, immutability, and thread safety

Mutability is a source of bugs in multi-threaded applications. A mutable bug is normally a
data bug caused by values being updated and shared between threads. To remove the risk
of mutability bugs, it is best to use immutable types. The guaranteed safe execution of a
body of code by multiple threads at the same time is called thread safety. When working
with multi-threaded programs, it is important that your code is thread-safe. Your code is
thread-safe if it removes race conditions and deadlocks, along with problems caused by
mutability.

An object that cannot be modified after it has been created is an immutable object. Once
created, if passed between threads using correct thread synchronization, all threads will see
the same valid state of an object. Immutable objects allow you to share data safely between
threads.

An object that can be modified after it has been created is a mutable object. Mutable objects
can have their data values changed between threads. This can lead to some serious data
corruption. So, even if the program does not crash, it can leave the data in an invalid state.
Therefore, when working with multiple threads of execution, it is important that your
objects are immutable. In Chapter 3, Classes, Objects, and Data Structures, we went through
creating and using immutable data structures for your immutable objects.

To ensure thread safety, do not use mutable objects, pass parameters by

reference, or modify member variables—only pass parameters by value
and only operate on parameter variables. Do not access member variables.
Immutable structures are a good and thread-safe way to pass data
between objects.

We will take a brief look at mutability, immutability, and thread safety with the following
examples. We'll start with mutability in terms of thread safety.

Writing code that is mutable and not thread-safe

To demonstrate mutability within a multi-threaded application, we will write a new .NET
Framework console application. Add a new class to the application called MutableClass
with the following code:
internal class MutableClass
{
private readonly int[] _intArray;

public MutableClass(int[] intArray)

[ 234 ]
Threading and Concurrency Chapter 8

{
_intArray = intArray;
}

public int[] GetIntArray()

{
return _intArray;
}
}

In our MutableClass class, we have a constructor that takes an integer array as an

argument. A member integer array is then assigned the array passed into the constructor.
The GetIntArray() method returns the integer array member variable. If you look at this
class, you would not think it is mutable because once the array is passed into the
constructor, the class provides no way to modify it. Yet, the integer array passed into the
constructor is mutable. The GetIntArray() method returns a reference to the mutable
array.

In our Program class, we will add the MutableExample() method to show that the integer
array is mutable:
private static void MutableExample()
{
int[] iar = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 };
var mutableClass = new MutableClass(iar);

Console.WriteLine($"Initial Array: {iar[0]}, {iar[1]}, {iar[2]},

{iar[3]}, {iar[4]}, {iar[5]}, {iar[6]}, {iar[7]}, {iar[8]}, {iar[9]}");

for (var x = 0; x < 9; x++)

{
var thread = new Thread(() =>
{
iar[x] = x + 1;
var ia = mutableClass.GetIntArray();
Console.WriteLine($"Array [{x}]: {ia[0]}, {ia[1]}, {ia[2]},
{ia[3]}, {ia[4]}, {ia[5]}, {ia[6]}, {ia[7]}, {ia[8]}, {ia[9]}");
});
thread.Start();
}
}

[ 235 ]
Threading and Concurrency Chapter 8

In our MutableExample() method, we have declared and initiated an integer array of

items from 0 to 9. We then declare a new instance of MutableClass and pass in the integer
array. Next, we print out the contents of the initial array before it is modified. Then, we
loop nine times. For each iteration, we increase the array at the index specified by the
current loop count value of x so that it equals x + 1. After that, we start the thread. Now,
update the Main() method, as follows:
static void Main(string[] args)
{
MutableExample();
Console.ReadKey();
}

Our Main() method simply calls MutableExample() and then waits for a keypress. Run
the code and you should see something as in the following screenshot:

As you can see, even though we only created one instance of MutableClass before
creating and running our threads, changing the local array modifies the array in the
instance of MutableClass. This proves that the arrays are mutable, and so they are not
thread-safe.

We will now look at immutability in terms of thread safety.

Writing code that is immutable and thread-safe

In our immutability example, we will again create a .NET Framework console application
and we'll use the same array. Add a class called ImmutableStruct and modify the code,
as shown:
internal struct ImmutableStruct
{

[ 236 ]
Threading and Concurrency Chapter 8

private ImmutableArray<int> _immutableArray;

public ImmutableStruct(ImmutableArray<int> immutableArray)

{
_immutableArray = immutableArray;
}

public int[] GetIntArray()

{
return _immutableArray.ToArray<int>();
}
}

Instead of using a normal integer array, we employ ImmutableArray. An immutable array

is passed into the constructor and assigned to the _immutableArray member variable. Our
GetIntArray() method returns the immutable array as a normal integer array.

Add the ImmutableExample() array to the Program class:

private static void ImmutableExample()
{
int[] iar = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 };
var immutableStruct = new ImmutableStruct(iar.ToImmutableArray<int>());

Console.WriteLine($"Initial Array: {iar[0]}, {iar[1]}, {iar[2]},

{iar[3]}, {iar[4]}, {iar[5]}, {iar[6]}, {iar[7]}, {iar[8]}, {iar[9]}");

for (var x = 0; x < 9; x++)

{
var thread = new Thread(() =>
{
iar[x] = x + 1;
var ia = immutableStruct.GetIntArray();
Console.WriteLine($"Array [{x}]: {ia[0]}, {ia[1]}, {ia[2]},
{ia[3]}, {ia[4]}, {ia[5]}, {ia[6]}, {ia[7]}, {ia[8]}, {ia[9]}");
});
thread.Start();
}
}

In our ImmutableExample() method, we create an array of integers that we pass into the
constructor of ImmutableStruct as an immutable array. We then print the content of the
local array before modification. Then, we loop nine times. In each iteration, we access the
location of the count of the current iteration in the array and add the count of the current
iteration plus one to the variable at that position in the array.

[ 237 ]
Threading and Concurrency Chapter 8

We then assign a copy of the immutableStruct array to a local variable via a call to
GetIntArray(). Then, we proceed to print out the values of the returned array. Finally,
we start the thread. Call the ImmutableExample() method from your Main() method and
then run the code. You should see the following output:

As you can see, the array's content is not modified by updating the local array. This version
of our program shows that our program is thread-safe.

Let's briefly run through what we've learned about thread safety so far in the next section.

Understanding thread safety

As you saw in the previous two sections, it is very important to be careful when writing
multi-threaded code. Writing thread-safe code can be very difficult, especially in larger
projects. You have to be particularly careful with collections, passing parameters by
reference, and when accessing member variables within static classes. The best practices for
multi-threaded applications are to only pass immutable types, not to access static member
variables, and if any code that is not thread-safe must be executed, then to lock the code
using a lock, mutex, or semaphore. Although you have already seen code like this in action
in this chapter, we will quickly refresh our memory on this with some code snippets.

[ 238 ]
Threading and Concurrency Chapter 8

The following code snippet shows how to write an immutable type using readonly
struct:

public readonly struct ImmutablePerson

{
public ImmutablePerson(int id, string firstName, string lastName)
{
_id = id;
_firstName = firstName;
_lastName = lastName;
}

public int Id { get; }

public string FirstName { get;
public string LastName { get { return _lastName; } }
}

In our ImmutablePerson structure, we have a public constructor that takes an integer for
the ID and strings for the first and last name. We assign the id, firstName, and
lastName parameters to member read-only variables. The only access to the data is via
read-only properties. This means that there is no way to modify the data. Since the data
cannot be modified once it has been created, it is classed as thread-safe. Because it is thread-
safe, it cannot be modified by different threads. The only way to modify the data would be
to create a new struct with the new data.

Structs can be mutable, just like classes. However, to pass data around
that you don't want to be modified, then read-only structs are a good,
lightweight choice. They are faster to create and destroy than classes as
they are added to the stack—that is, unless they are part of a class that is
added to the heap.

Earlier on, we saw how collections are mutable. However, there is also a namespace of
immutable collections called System.Collections.Namespace. The following table lists
various items from this namespace:

[ 239 ]
Threading and Concurrency Chapter 8

The System.Collections.Immutable namespace contains a number of

immutable collections that you can use safely between threads. Refer
to https://docs.microsoft.com/en-us/dotnet/api/system.
collections.immutable?view=netcore-3.1 for more details.

Using a lock object in C# is really straightforward, as the following code snippet shows:
public class LockExample
{
public object _lock = new object();

public void UnsafeMethod()

{
lock(_lock)
{

[ 240 ]
Threading and Concurrency Chapter 8

// Execute unsafe code.

}
}
}

We create and instantiate the _lock member variable. Then, when it comes to executing
code that is not thread-safe, we wrap the code in the lock and pass in the _lock variable to
use as the lock object. When a thread enters the lock, all other threads are barred from
executing the code until the thread leaves the lock. One problem with using this code is that
threads can enter a deadlock situation. One way around this is to use a mutex.

You can use a synchronization primitive for interprocess synchronization. Start by adding
the following code to the top of the class that has code that needs protection:
private static readonly Mutex _mutex = new Mutex();

Then, to use the mutex, you will need to wrap the code that needs protection with the
following try/catch block:
try
{
_mutex.WaitOne();
// ... Do work here ...
}
finally
{
_mutex.ReleaseMutex();
}

In the preceding code, the WaitOne() method blocks the current thread until the wait
handle receives a signal. As soon as the mutex is signaled, the WaitOne() method returns
true. The calling thread then assumes ownership of the mutex. Protected resources can
then be accessed by the calling thread. When the work is finished on the protected resource,
the mutex is released by calling ReleaseMutex(). ReleaseMutex() is called in the
finally block because you don't want a thread to keep a resource locked if it raises an
exception for whatever reason. So, always release a mutex in a finally block.

[ 241 ]
Threading and Concurrency Chapter 8

Another mechanism for protecting access to resources is using a semaphore. Semaphores

are coded much like a mutex and they perform the same role of protecting resources. The
main difference between a semaphore and a mutex is that a mutex is a locking mechanism
and a semaphore is a signaling mechanism. To use semaphores instead of locks and
mutexes, add the following line to the top of a class:
private static readonly Semaphore _semaphore = new Semaphore(2, 4);

We have now added a new semaphore variable. The first parameter states the initial
number of requests for the semaphore that can be granted concurrently. The second
parameter states the maximum number of requests for the semaphore that can be granted
concurrently. You will then protect access to a resource in your methods, as follows:
try
{
_semaphore.WaitOne();
// ... Do work here ...
}
finally
{
_semaphore.Release();
}

The current thread is blocked until the current wait handle receives a signal. The thread can
then do its work. Finally, the semaphore is released.

You have seen, in this chapter, how to use locks, mutexes, and semaphores to lock code that
is not thread-safe. Also remember that background threads terminate when the process
completes and terminates, whereas foreground threads will continue executing until
completion. If you have any code that must run to completion without the thread being
terminated halfway through what it is doing, then you are better off using foreground
threads over background threads.

The next section covers synchronized method dependencies.

Synchronized method dependencies

To synchronize your code, use a lock statement as we did previously. You can also
reference the System.Runtime.CompilerServices namespace in your projects. Then,
you can add the [MethodImpl(MethodImplOptions.Synchronized)] annotation to
methods and properties.

[ 242 ]
Threading and Concurrency Chapter 8

Here is an example of
the [MethodImpl(MethodImplOptions.Synchronized)] annotation applied to a
method:
[MethodImpl(MethodImplOptions.Synchronized)]
public static void ThisIsASynchronisedMethod()
{
Console.WriteLine("Synchronised method called.");
}

Here is an example of using [MethodImpl(MethodImplOptions.Synchronized)] with

a property:
private int i;
public int SomeProperty
{
[MethodImpl(MethodImplOptions.Synchronized)]
get { return i; }
[MethodImpl(MethodImplOptions.Synchronized)]
set { i = value; }
}

As you can see, it is easy to encounter a deadlock or a race condition, but it is just as easy to
overcome deadlocks by using Monitor.TryEnter() and race conditions
with Task.ContinueWith().

In the next section, we look at the Interlocked class.

Using the Interlocked class

In multi-threaded applications, errors can creep in during the thread scheduler context-
switching process. One of the main problems that arises is the update of the same variables
by different threads. The methods of the System.Threading.Interlocked class in the
mscorlib assembly help to protect against these kinds of errors. The methods of the
Interlocked class do not throw exceptions, and so they are very helpful in applying
simple state changes in a more performant way than using the lock statement that we've
seen previously.

[ 243 ]
Threading and Concurrency Chapter 8

The methods available in the Interlocked class are as follows:

CompareExchange: Compares two variables and stores the results in a different

variable
Add: Adds two Int32 or Int64 integer variables together and stores the result in
the first integer
Decrement: Decrements the Int32 and Int64 integer variable values and stores
their results
Increment: Increments the Int32 and Int64 integer variable values and stores
their results
Read: Reads integer variables of the Int64 type
Exchange: Exchanges values between variables

We are now going to write a simple console application that demonstrates these methods.
Start by creating a new .NET Framework console application. Add the following lines to the
top of the Program class:
private static long _value = long.MaxValue;
private static int _resourceInUse = 0;

The _value variable will be used to demonstrate the update of variables using the
interlocking methods. The _resourceInUse variable is used to indicate whether a resource
is in use. Add the CompareExchangeVariables() method:
private static void CompareExchangeVariables()
{
Interlocked.CompareExchange(ref _value, 123, long.MaxValue);
}

In our CompareExchangeVariables() method, we call the CompareExchange() method

to compare _value with long.MaxValue. If the two values are equal, then _value is
replaced with the value of 123. We'll now add our AddVariables() method:
private static void AddVariables()
{
Interlocked.Add(ref _value, 321);
}

[ 244 ]
Threading and Concurrency Chapter 8

The AddVariables() method calls the Add() method to access the _value member
variable and update it with the value of _value plus 321. Next, we'll add our
DecrementVariable() method:

private static void DecrementVariable()

{
Interlocked.Decrement(ref _value);
}

This method calls the Decrement() method, which decrements the _value member
variable by 1. Our next method is IncrementValue():
private static void IncrementVariable()
{
Interlocked.Increment(ref _value);
}

In our IncrementVariable() method, we increment the _value member variable by

calling the Increment() method. The next method we will write is the ReadVariable()
method:
private static long ReadVariable()
{
// The Read method is unnecessary on 64-bit systems, because 64-bit
// read operations are already atomic. On 32-bit systems, 64-bit read
// operations are not atomic unless performed using Read.
return Interlocked.Read(ref _value);
}

Since 64-bit read operations are atomic, calling the Interlocked.Read() method is
unnecessary. However, on 32-bit systems, for 64-bit reads to be atomic, you need to call the
Interlocked.Read() method. Add the PerformUnsafeCodeSafely() method:

private static void PerformUnsafeCodeSafely()

{
for (int i = 0; i < 5; i++)
{
UseResource();
Thread.Sleep(1000);
}
}

[ 245 ]
Threading and Concurrency Chapter 8

The PerformUnsafeCodeSafely() method loops five times. Each iteration of the loop
calls the UseResource() method, and then the thread goes to sleep for 1 second. Now,
we'll add the UseResource() method:
static bool UseResource()
{
if (0 == Interlocked.Exchange(ref _resourceInUse, 1))
{
Console.WriteLine($"{Thread.CurrentThread.Name} acquired the
lock");
NonThreadSafeResourceAccess();
Thread.Sleep(500);
Console.WriteLine($"{Thread.CurrentThread.Name} exiting lock");
Interlocked.Exchange(ref _resourceInUse, 0);
return true;
}
else
{
Console.WriteLine($"{Thread.CurrentThread.Name} was denied the
lock");
return false;
}
}

The UseResource() method prevents a lock from being obtained if the resource is in use,
as identified by the _resourceInUse variable. We start by setting the _resourceInUse
member variable value to 1 by calling the Exchange() method. The Exchange() method
returns an integer, which we compare against 0. If the value returned by Exchange() is 0,
then the method is not in use.

If the method is in use, then we output a message informing the user that the current thread
was denied the lock.

If the method is not in use, then we output a message informing the user that the current
thread has obtained a lock. We then call the NonThreadSafeResourceAccess() method
and then send the thread to sleep for half a second to simulate work.

When the thread awakes, we output a message informing the user that the current thread
has exited the lock. Then, we release the lock by calling the Exchange() method and
setting the value of _resourceInUse to 0. Add the NonThreadSafeResourceAccess()
method:
private static void NonThreadSafeResourceAccess()
{
Console.WriteLine("Non-thread-safe code executed.");
}

[ 246 ]
Threading and Concurrency Chapter 8

NonThreadSafeResourceAccess() is where non-thread-safe code is executed in the

safety of the lock. In our method, we simply inform the user with a message. The last job to
do before we run our code is to update our Main() method, as follows:
static void Main(string[] args)
{
CompareExchangeVariables();
AddVariables();
DecrementVariable();
IncrementVariable();
ReadVariable();
PerformUnsafeCodeSafely();
}

Our Main() method calls the methods that test the Interlocked methods. Run the code
and you should see something similar to the following:

We'll now go over some general recommendations.

General recommendations
In this final section, we will look at some general recommendations from Microsoft for
working on multi-threaded applications. They include the following:

Avoid using Thread.Abort to terminate other threads.

Use a mutex, ManualResetEvent, AutoResetEvent, and Monitor to
synchronize activities between multiple threads.

[ 247 ]
Threading and Concurrency Chapter 8

Where possible, use a thread pool for your worker threads.

If you have any worker threads that gets blocked, then use Monitor.PulseAll
to notify all the threads of a change in the worker thread's state.
Avoid using this, type instances, and string instances including string literals as
lock objects. Avoid using types of the lock objects.
Instance locks can result in deadlocks, so exercise caution when using them.
Use the try/finally block with threads that enter a monitor so that in the
finally block, you ensure that the thread leaves the monitor by calling
Monitor.Exit().
Use different threads for different resources.
Avoid assigning multiple threads to the same resource.
I/O tasks should have their own thread as they block when performing I/O
operations. This way, you allow other threads to run.
User input should have its own dedicated thread.
Improve performance for simple state changes by using the methods of the
System.Threading.Interlocked class instead of the lock statement.
For heavily used code, avoid synchronization as it can lead to deadlocks and race
conditions.
Make static data thread-safe by default.
Instance data must not be thread-safe by default; otherwise, you decrease
performance, increase lock contention, and introduce the possibility of race
conditions and deadlocks occurring.
Avoid using static methods that alter state as they lead to threading bugs.

That concludes our look at threading and concurrency. Let's run through a summary of
what we have learned.

Summary
In this chapter, we covered what threading is and how to use it. We looked at the problems
of deadlocks and race conditions in action, and we saw how to prevent these exceptional
circumstances using a lock statement and the TPL library. We also discussed the thread
safety of static constructors, static methods, immutable objects, and mutable objects. We
saw why using immutable objects is a thread-safe way of transferring data between
threads, and we reviewed some general recommendations for working with threads.

[ 248 ]
Threading and Concurrency Chapter 8

We also saw how making your code thread-safe can have a lot of benefits. In the next
chapter, we will look at designing effective APIs. But for now, you can test your knowledge
by answering the following questions and you can further your reading by referring to the
links provided.

Questions
1. What is a thread?
2. How many threads are there in a single-threaded application?
3. What types of threads are there?
4. What thread terminates as soon as the program is exited?
5. What thread continues through to completion, even if the program is exited?
6. What code makes a thread sleep for half a millisecond?
7. How do you instantiate a thread that calls a method named Method1?
8. How do you make a thread a background thread?
9. What is a deadlock?
10. How do you exit a lock obtained using Monitor.TryEnter(objectName)?
11. How can you recover from a deadlock?
12. What is a race condition?
13. What is one way to prevent race conditions?
14. What makes static methods unsafe?
15. Are static constructors thread-safe?
16. What is responsible for managing groups of threads?
17. What is an immutable object?
18. Why are immutable objects preferred to mutable objects in threaded
applications?

Further reading
https://www.c-sharpcorner.com/blogs/mutex-and-semaphore-in-
thread provides examples of using a mutex and a semaphore.
https://www.guru99.com/mutex-vs-semaphore.html explains the differences
between a mutex and a semaphore.

[ 249 ]
Threading and Concurrency Chapter 8

https://docs.microsoft.com/en-us/dotnet/csharp/programming-guide/
classes-and-structs/static-constructors is the official Microsoft
documentation on static constructors.
https://docs.microsoft.com/en-us/dotnet/standard/threading/managed-
threading-best-practices is the official Microsoft guidance on Microsoft's
managed threading best practices.
https://docs.microsoft.com/en-us/dotnet/standard/parallel-programming/
task-parallel-library-tpl is the official Microsoft API documentation for the
TPL.
https://www.c-sharpcorner.com/UploadFile/1d42da/interlocked-class-in-
C-Sharp-threading/ covers the Interlocked class in C# threading.
http://geekswithblogs.net/BlackRabbitCoder/archive/2012/08/23/c.net-
little-wonders-interlocked-read-and-exchange.aspx provides a discussion
on System.Threading.Interlocked with examples.
http://www.albahari.com/threading/ is a link to a free eBook by Joseph
Albahari about threading in C#.
https://docs.microsoft.com/en-us/dotnet/api/system.collections.
immutable?view=netcore-3.1 is the official Microsoft documentation on the
immutable collections available in the System.Collections.Immutable
namespace.

[ 250 ]
9
Designing and Developing APIs
Application Programming Interfaces (APIs) have never been so vital in so many ways as
they are these days. APIs are used to connect governments and institutions in the sharing of
data and in a collaborative manner for business and governmental issues. They are used
between doctors' surgeries and hospitals to share patient data in real time. You use APIs
every day when you connect to your emails and collaborate with your colleagues and
clients through platforms such as Microsoft Teams, Microsoft Azure, Amazon Web
Services, and Google Cloud Platform.

Every time you chat with someone or have a video call with them using your computers or
phones, you are using APIs. When streaming video conferences, entering a website
technical support chat, or streaming your favorite music and videos, you are using APIs.
So, as a programmer, it is imperative that you are well versed in what APIs are and how to
design, develop, secure, and deploy them.

In this chapter, we will talk about what APIs are, how they benefit you, and why it is
necessary to learn about them. We will also be discussing API proxies, design and
development guidelines, how to design APIs using RAML, and how to document APIs
using Swagger.

The following topics are covered in this chapter:

What is an API?
API proxies
API design guidelines
API design using RAML
Swagger API development
Designing and Developing APIs Chapter 9

This chapter will assist you in gaining the following skills:

Understanding APIs and why you need to learn about them

Understanding API proxies and why we use them
Being aware of design guidelines when designing your own APIs
Using RAML to design your own APIs
Using Swagger to document your APIs

By the end of this chapter, you will understand the basics of good API design and you will
be armed with the knowledge needed to push your API abilities forward. It is important to
understand what an API is, and so that is how we shall start this chapter. But first, make
sure that you implement the following technical requirements to get the most out of this
chapter.

Technical requirements
We will be using the following technologies in this chapter to create an API:

Visual Studio 2019 Community edition or higher

Swashbuckle.AspNetCore 5 or higher
Swagger (https://swagger.io)
Atom (http://atom.io)
API Workbench by MuleSoft

What is an API?
APIs are reusable libraries that can be shared between different applications and can be
made available via REST services (in which case, they are referred to as RESTful APIs).

Representational State Transfer (REST) was introduced by Roy Fielding

in 2000.

[ 252 ]
Designing and Developing APIs Chapter 9

REST is an architectural style that is made up of constraints. Altogether there are six
constraints that should be considered when writing REST services. These constraints are as
follows:

Uniform interface: This is used to identify resources, and it manipulates these

resources through representation. Messages use hypermedia and are self-
descriptive. Hypermedia as the Engine of Application State (HATEOAS) is
utilized to contain information about what operation can be carried out next by
the client.
Client-server: This constraint utilizes information hiding through
encapsulation. So, only the API calls that are to be used by clients will be visible
and all the other APIs will be kept hidden. A RESTful API should be
independent of other parts of the system, making it loosely coupled.
Stateless: This states that the RESTful API has no session nor history. If a session
or history is required by the client, then the client must provide all the relevant
information in the request to the server.
Cacheable: This constraint means that resources must declare themselves
cacheable. This means that resources can be accessed quickly. As a result, our
RESTful API gains speed and our server load is reduced.
Layered system: The layered system constraint dictates that each layer must do
one and only one thing. Each component should only know what it needs to use
in order to function and perform its tasks. A component should not know about
the parts of the system that it does not use.
Optional executable code: The executable code constraint is optional. This
constraint determines that servers can, on a temporary basis, extend or customize
the functionality of a client by transferring executable code.

So, when designing an API, it would be prudent to assume that the end user will be a
programmer with any level of experience. They should be able to easily obtain the API,
read up on it, and put it to work straight away.

Don't worry about creating the perfect API. APIs usually evolve over time anyway, and if
you have ever worked with Microsoft APIs, you will know that they regularly upgrade
them. APIs with features that will be removed in the future are often marked with an
annotation that informs the user not to use a particular property or method as they will be
removed in a future release. Then, when they will no longer be used, they usually get
marked with an obsolete annotation before they are finally removed. This tells the users of
the API to upgrade any apps using deprecated features.

[ 253 ]
Designing and Developing APIs Chapter 9

Why use REST services for API access? Well, many companies make large profits from
making their APIs available online and charging for them. So, RESTful APIs can be a very
valuable asset. Rapid API (https://rapidapi.com/) has free and paid APIs available for
use.

Your APIs can remain permanently in place. If you use a cloud provider, your APIs can be
highly scalable and you can make them generally available either for free or via a
subscription. You can encapsulate all the complicated workings and expose what you need
to via a simple interface, and because your APIs will be small and cacheable, they are very
fast. Let's look now at API proxies and why you would use them.

API proxies
An API proxy is a class that sits between the client and your API. It is, in essence, an API
contract between you and the developers who will be using your API. So, rather than
giving developers direct access to your API's backend services, which may break over time
as you refactor and extend them, you provide assurance to the consumers of your API that
the API contract will be honored, even when the backend services change.

The following diagram displays the communication between the client, an API proxy, the
actual API being accessed, and the API's communication with the data source:

A console application that shows how easy it is to implement the proxy pattern will be
programmed in this section. Our example will have an interface that will be implemented
by the API and the proxy. The API will return the actual message and the proxy will obtain
the message from the API and pass it to the client. Proxies can also do much more than
simply call the API method and return a response. They can perform authentication,
authorization, routing based on credentials, and much more. However, our example will be
kept to the absolute minimum so that you can see the simplicity in the proxy pattern.

[ 254 ]
Designing and Developing APIs Chapter 9

Start a new .NET Framework console application. Add the Apis, Interfaces,
and Proxies folders and place the HelloWorldInterface interface into
the Interfaces folder:
public interface HelloWorldInterface
{
string GetMessage();
}

Our interface method, GetMessage(), returns a message as a string. Both the proxy and
API class will implement this interface. The HelloWorldApi class implements
HelloWorldInterface, so add it to the Apis folder:

internal class HelloWorldApi : HelloWorldInterface

{
public string GetMessage()
{
return "Hello World!";
}
}

As you can see, our API class implements the interface and returns a "Hello
World!" message. We have also made the class an internal class. This prevents external
callers from being able to access the contents of this class. Now, we'll add
our HelloWorldProxy class to the Proxies folder:
public class HelloWorldProxy : HelloWorldInterface
{
public string GetMessage()
{
return new HelloWorldApi().GetMessage();
}
}

Our proxy class is set to public as this class will be called by clients. The proxy class will
call the GetMessage() method within the API class and return the response to the caller.
All that's left to do now is to modify our Main() method:
static void Main(string[] args)
{
Console.WriteLine(new HelloWorldProxy().GetMessage());
Console.ReadKey();
}

[ 255 ]
Designing and Developing APIs Chapter 9

Our Main() class calls the GetMessage() method of the HelloWorldProxy proxy class.
Our proxy class calls the API class and the returned method is printed in the console
window. The console then waits for a keypress before it exits.

Run the code and view the output; you have successfully implemented an API proxy class.
You can make your proxies as simple or as complicated as they need to be, but what you
have done here is the foundation for success.

In this chapter, we will be building an API. So, let's discuss what we will be building, and
then get going with working on it. Once you've completed the project, you will have a
working API that generates a monthly dividend payment calendar in JSON format.

API design guidelines

There are some basic guidelines to follow to write an effective API—for example, your
resources should use nouns in plural form. So, for example, if you had a wholesale website,
then your URLs would look something like the following dummy links:

http://wholesale-website.com/api/customers/1
http://wholesale-website.com/api/products/20

The preceding URLs will follow the controller routes of api/controller/id. In terms of
relationships within the business domain, these should also be reflected in URLs such as
http://wholesale-website.com/api/categories/12/products—this call will
return a list of products for category 12.

If you need to use a verb as a resource, then you can do so. When making an HTTP request,
use GET to retrieve items, HEAD to retrieve only headers, POST to insert or save a new
resource, PUT to replace a resource, and DELETE to remove a resource. Keep resources lean
by using query parameters.

When paginating results, a ready-made set of links should be made available to the client.
RFC 5988 introduced link headers. In the specification, an International Resource
Identifier (IRI) is a typed connection between two resources. For more information, refer
to https://www.greenbytes.de/tech/webdav/rfc5988.html. The format of link header
requests is as follows:

<https://wholesale-
website.com/api/products?page=10&per_page=100>; rel="next"
<https://wholesale-
website.com/api/products?page=11&per_page=100>; rel="last"

[ 256 ]
Designing and Developing APIs Chapter 9

The versioning of your API can be done in the URL. So, each resource will have a different
URL for the same resource, as in the following examples:

https://wholesale-website.com/api/v1/cart
https://wholesale-website.com/api/v2/cart

This form of versioning is very simple and makes it easy to find the correct version of the
API.

JSON is the preferred resource representation. It is much more human-readable than XML
and is also lighter in size. When you are using the POST, PUT, and PATCH verbs, you should
also require the content-type header to be set to application/JSON, or throw the 415 HTTP
status code (which means unsupported media type). Gzip is a single-file/stream lossless
data compression utility. Use Gzip by default to save a good percentage in bandwidth, and
always set the HTTP Accept-Encoding header to gzip.

Always use HTTPS (TLS) for your APIs. The identification of the caller should always be
done in the header. We saw this with our API when we set the x-api-key header with our
API access key. Each request should be authenticated and authorized. Unauthorized access
should result in an HTTP 403 Forbidden response. Also, use the correct HTTP response
codes. So, if a request is successful, use the 200 status code, for a resource that is not found,
use 404, and so on. For an exhaustive list of HTTP status codes, visit https://
httpstatuses.com/. OAuth 2.0 is the industry-standard protocol for authorization. You can
read all about it at https://oauth.net/2/.

An API should provide documentation on its usage with examples. Documentation should
always be up to date with the current version, and it should be visually appealing and easy
to read. We'll look at Swagger to help us create documentation later in this chapter.

You never know when your API is going to need to scale. So, this should be factored in
from the start. In our Dividend Calendar API project in the next chapter, you will see how we
implement throttling to just one API call per month on a specific day of the month.
However, you can effectively come up with 1,001 different ways to throttle your APIs
depending on your own needs, but this should be done at the start of a project. So, as soon
as you start a new project, think scalability.

For security and performance reasons, you may decide to implement an API proxy. An API
proxy disconnects the client from accessing your API directly. A proxy can access an API in
the same project or on an external API. By using a proxy, you can avoid exposing your
database schema.

[ 257 ]
Designing and Developing APIs Chapter 9

Responses to the client should never match the structure of your database. This can act as a
green light for hackers. So, avoid one-to-one mappings between database structures and the
responses you send back to clients. You should also hide identifiers from your clients as
they can be used to manually access data by the client.

An API contains resources. A resource is an item that can be operated on in some way.
Resources can be files or data. For example, students in a school database are resources that
can be added, edited, or deleted. Video files can be retrieved and played, as can audio files.
Images are also resources, as are report templates that will be opened, manipulated, and
filled with data before they are presented to the user.

Often, resources form collections of items, such as students in a school database. Students
is the name of a collection of the Student type. Resources are accessed via a URL. A URL
contains the path to a resource.

URLs are known as API endpoints. An API endpoint is an address of a resource. This
resource may be accessed by an URL with one or more parameters or an URL without any
parameters. An URL should only contain plural nouns (names of resources) and should not
contain verbs or actions. Parameters can be used to identify a single resource within a
collection. Pagination should be employed if the dataset is going to be very large. For
requests with parameters that break the URI length limit, you can place the parameters in
the body of a POST request.

Verbs form part of the HTTP request. The POST verb is used to add a resource. To retrieve
one or more resources, you use the GET verb. PUT updates or replaces one or more
resources, and PATCH updates or modifies a resource or collection. DELETE deletes a
resource or collection.

You should always make sure you provide and respond to HTTP status codes
appropriately. For a complete list of HTTP status codes, visit https://httpstatuses.com/.

As for field, method, and property names, you can use any convention you like, but it must
be consistent and follow the company guidelines. Camel case convention is normally used
in JSON. Since you will be developing APIs in C#, it is best to stick to the industry-standard
C# naming conventions.

Since your API will evolve over time, it is best to employ some form of versioning.
Versioning allows consumers to consume specific versions of your API. This can be very
important for providing backward compatibility when new versions of your API
implement breaking changes. It is normally a good idea to have the version number, such
as v1 or v2, included in the URL. Whatever method you use to version your APIs, just
remember to be consistent.

[ 258 ]
Designing and Developing APIs Chapter 9

If you will be consuming third-party APIs, you will need to keep the API keys secret. One
way to accomplish this is to store the keys in a key vault, such as Azure Key Vault, which
requires authentication and authorization. You should also secure your own APIs with a
method of your choosing. A common method nowadays is through the use of API keys.
You will see how to use API keys and Azure Key Vault to secure third-party keys and your
own APIs in the next chapter.

Well-defined software boundaries

Nobody in their right mind likes spaghetti code. It is very hard to read, maintain, and
extend. So, when designing an API, you can overcome this problem with well-defined
software boundaries. A well-defined software boundary is known as a bounded context in
Domain-Driven Design (DDD). In business terms, a bounded context is a business
operational unit, such as HR, finance, customer services, infrastructure, and so on. These
business operational units are known as domains, and they can be broken down into
smaller sub-domains. Then, these sub-domains can be broken down into even smaller sub-
domains.

By breaking a business up into business operational units, domain experts can be employed
in those specific areas. A common language can be determined at the start of a project so
that the business understands the IT terms and the IT staff understands the business terms.
If the business and IT staff are language-aligned, there is less margin for errors due to
misunderstandings from both sides.

Having a major project broken down into sub-domains means that you can have smaller
teams working independently on projects. So, large development teams can be grouped
into smaller teams working concurrently on various projects.

DDD is a big subject in itself and is not covered here. However, links to more information
are posted in the Further reading section of this chapter.

The only items that should be exposed by APIs are the interfaces that form contracts and
API endpoints. Everything else should be hidden from the subscriber and consumer. This
means that even large databases can be broken down so that each API has its own database.
Given how large and complex websites can be by today's standards, we can even have
micro-services with micro-databases and micro-frontends.

A micro-frontend is a small portion of a web page that is dynamically retrieved and

modified according to user interactions. That frontend will interact with an API, which in
turn will access a micro-database. This is ideal in terms of Single-Page Applications
(SPAs).

[ 259 ]
Designing and Developing APIs Chapter 9

SPAs are websites consisting of a single page. When a user initiates an action, only the
required portion of the web page is updated; the rest of the page remains the same. So, for
example, say the web page has an aside. That aside displays adverts. Those adverts are
stored in the database as portions of HTML. The aside is set to auto-update itself every 5
seconds. When 5 seconds is up, the aside requests the API to assign a new advert. The API
then uses whatever algorithm is in place to obtain a new advert to display from the
database. The HTML document is then updated, and the aside is updated with the new
advert. The following diagram shows the typical SPA life cycle:

This aside is a well-defined software boundary. It does not need to know anything
whatsoever about the rest of the page it is displayed in. All it is concerned with is
displaying a new advert every 5 seconds:

The previous diagram shows an SPA communicating with a RESTful API via an API proxy,
and the API is able to access documents and databases.

[ 260 ]
Designing and Developing APIs Chapter 9

The only components that make up the aside are an HTML document fragment, a micro-
service, and a database. These can be worked on by a small team in whatever technology
that the team prefers and is comfortable with. The full SPA could be made up of hundreds
of micro-documents, micros-services, and micro-databases. The key point is that these
services could be made up of any technology and worked on independently by any team.
Multiple projects could also be worked on concurrently.

Within our bounded context, we can use the following software methodologies to improve
the quality of our code:

The Single Responsibility, Open/Closed, Liskov, Interface Segregation, and

Dependency Inversion (SOLID) principles
Don't Repeat Yourself (DRY)
You Ain't Gonna Need It (YAGNI)
Keep It Simple, Stupid (KISS)

These methodologies work well together to eliminate duplicate code, prevent you from
writing code that you don't need, and to keep objects and methods small. The reason why
we develop for a class and a method is that they should both do only one thing and do it
well.

Namespaces are used to perform logical groupings. We can use namespaces to define
software boundaries. The more specific a namespace is, the more meaningful it is to the
programmer. Meaningful namespaces help programmers to partition code and find what
they are looking for with ease. Use namespaces to logically group interfaces, classes,
structs, and enums.

In the next section, you will learn how to design an API using RAML. Then, you will
generate a C# API from the RAML file.

Understanding the importance of good quality

API documentation
When working on a project, it is necessary to understand all the APIs that are already used.
The reason for this is that you can often end up writing code that already exists, which
obviously leads to wasted effort. Not only that but by writing your own version of code
that already exists, you now have two copies of code that do the same thing. This adds to
the complexity of the software and increases the maintenance overhead as both versions of
the code must be maintained. It also adds the potential for bugs.

[ 261 ]
Designing and Developing APIs Chapter 9

On massive projects that are spread across multiple technologies and repositories, with
teams that have a high staff turnaround, and especially where no documentation exists,
code duplication becomes a real problem. Sometimes, there will only be one or two domain
experts, with the majority of the team not knowing the system at all. I have worked on
projects like this before and they are a real pain to maintain and expand.

That is why API documentation is vital for any project, no matter how large or how small it
is. It is inevitable in the field of software development that people will move on, especially
when more lucrative work is offered elsewhere. If the person moving on is the domain
expert, then they will take their knowledge with them. If no documentation exists, then
new developers to the project will have a steep learning curve in understanding the project
by having to read the code. If the code is messy and complex, this can cause a real headache
for onboarding new staff.

As a result, due to the lack of system knowledge, programmers will be inclined to more or
less write the code they need from scratch to get the job done as they will be under pressure
to deliver to the business on time. This will often lead to duplicate code and code reuse not
being utilized. This causes the software to become complex and error-prone, and this kind
of software ends up being hard to extend and maintain.

Now, you understand why APIs must be documented. A well-documented API will lead to
greater understanding by programmers and is more inclined to get reused, thereby
reducing the potential for code duplication and producing code that is hard to extend or
maintain.

You should also be aware of any code that is marked as deprecated or obsolete. Deprecated
code will be removed in future releases and obsolete code is no longer in use. If you are
using APIs that are marked as deprecated or obsolete, then this code should be prioritized
to address first.

Now that you understand the importance of good quality API documentation, we will look
at a tool called Swagger. Swagger is an easy-to-use tool for producing nice-looking, high-
quality API documentation.

Swagger API development

Swagger provides a powerful set of tools that are focused around API development. With
Swagger, you can do the following things:

Design: Design your API and model it to keep up with specification-based

standards.
Build: Build an API in C# that is stable and reusable.

[ 262 ]
Designing and Developing APIs Chapter 9

Document: Provide developers with documentation that they can interact with.
Test: Easily test your API.
Standardize: Apply constraints to your API architecture using your company
guidelines.

We are going to get Swagger up and running in our ASP.NET Core 3.0+ project. So, start by
creating the project in Visual Studio 2019. Select the Web API and No Authentication
settings. Before we continue, it is worth noting that Swagger automatically generates
aesthetically pleasing documentation that is functional. Very little code is required to set up
Swagger, which is why many modern APIs use it.

Before we can use Swagger, we first need to install support for it in our project. To install
Swagger, you must install version 5 or higher of
the Swashbuckle.AspNetCore dependency package. As of the time of writing, the version
available on NuGet is version 5.3.3. After the installation is complete, we need to add the
Swagger services that we will be using to the services collection. In our case, we will only
be using Swagger to document our API. In the Startup.cs class, add the following line to
the ConfigureServices() method:
services.AddSwaggerGen(swagger =>
{
swagger.SwaggerDoc("v1", new OpenApiInfo { Title = "Weather Forecast
API" });
});

In the code we've just added, the Swagger documenting service has been assigned to the
services collection. Our API version is v1 and our API title is Weather Forecast API. We
now need to update the Configure() method to add our Swagger middleware, as follows,
immediately after the if statement:
app.UseSwagger();
app.UseSwaggerUI(c =>
{
c.SwaggerEndpoint("/swagger/v1/swagger.json", "Weather Forecast API");
});

In our Configure() method, we are informing our app to use Swagger and the Swagger
UI, and we assign our Swagger endpoint for Weather Forecast API. Next, you will need
to install the Swashbuckle.AspNetCore.Newtonsoft NuGet dependency package
(version 5.3.3, as of the time of writing). Then, add the following line to
your ConfigureServices() method:
services.AddSwaggerGenNewtonsoftSupport();

[ 263 ]
Designing and Developing APIs Chapter 9

We added Newtonsoft support for our Swagger documentation generation. That is all there
is to getting Swagger up and running. So, run your project and navigate to
https://localhost:PORT_NUMBER/swagger/index.html. You should see the
following web page:

We will now take a look at why we should pass immutable structs instead of mutable
objects.

[ 264 ]
Designing and Developing APIs Chapter 9

Passing immutable structs instead of mutable

objects
In this section, you are going to write a computer program that processes 1 million objects
and 1 million immutable structs. You will see how much quicker, in terms of performance,
structs are over objects. We will be writing some code that processes 1 million objects in
1,440 milliseconds and processes 1 million structs in 841 milliseconds. That is a difference of
599 milliseconds. Such a small unit of time might not sound like a lot, but when working
with massive datasets, you will see big performance improvements when using immutable
structs over mutable objects.

Values in mutable objects can also be modified between threads, which can be very bad for
business. Imagine having £15,000 in your bank account and you pay your landlord £435 in
rent. Your account has an overdraft limit that can be exceeded. Now, at the same time that
you are paying £435, someone else is paying a car firm £23,000 for a new car. The value on
your account is modified by the car purchaser's thread. So, you end up paying your
landlord £23,000, leaving your bank balance £8,000 in debt. We won't code an example of
mutable data being changed between threads as this was covered in Chapter 8, Threading
and Concurrency.

The key points of this section are that structs are faster than objects and
immutable structs are thread-safe.

When creating and passing objects, structs are more performant than objects. You can also
make structs immutable so that they are thread-safe. Here, we will write a small program.
This program will have two methods—one will create 1 million person objects and the
other will create 1 million person structures.

Add a new .NET Framework console application

called CH11_WellDefinedBoundaries and the following PersonObject class:
public class PersonObject
{
public string FirstName { get; set; }
public string LastName { get; set; }
}

[ 265 ]
Designing and Developing APIs Chapter 9

This object will be used to create 1 million people objects. Now, add the PersonStruct:
public struct PersonStruct
{
private readonly string _firstName;
private readonly string _lastName;

public PersonStruct(string firstName, string lastName)

{
_firstName = firstName;
_lastName = lastName;
}

public string FirstName => _firstName;

public string LastName => _lastName;
}

This struct is immutable, with the readonly properties being set via the constructor and
used to create our 1 million structs. Now, we can modify the program to show the
performance between object and struct creation. Add the CreateObject() method:
private static void CreateObjects()
{
Stopwatch stopwatch = new Stopwatch();
stopwatch.Start();
var people = new List<PersonObject>();
for (var i = 1; i <= 1000000; i++)
{
people.Add(new PersonObject { FirstName = "Person", LastName =
$"Number {i}" });
}
stopwatch.Stop();
Console.WriteLine($"Object: {stopwatch.ElapsedMilliseconds}, Object
Count: {people.Count}");
GC.Collect();
}

As you can see, we start a stopwatch, create a new list, and add 1 million person objects to
the list. We then stop the stopwatch, output the results to the window, and then call the
garbage collector to clean up our resources. Let's now add our CreateStructs() method:
private static void CreateStructs()
{
Stopwatch stopwatch = new Stopwatch();
stopwatch.Start();
var people = new List<PersonStruct>();
for (var i = 1; i <= 1000000; i++)

[ 266 ]
Designing and Developing APIs Chapter 9

{
people.Add(new PersonStruct("Person", $"Number {i}"));
}
stopwatch.Stop();
Console.WriteLine($"Struct: {stopwatch.ElapsedMilliseconds}, Struct
Count: {people.Count}");
GC.Collect();
}

Our structure does a similar thing here as for the CreateObjects() methods, but creates a
list of structs and adds 1 million structs to the list. Finally, modify the Main() method, as
follows:
static void Main(string[] args)
{
CreateObjects();
CreateStructs();
Console.WriteLine("Press any key to exit.");
Console.ReadKey();
}

We call both of our methods and then wait for the user to press any key before we exit. Run
the program and you should see the following output:

As you can see from the previous screenshot, it took 1,440 milliseconds to create 1 million
objects and add them to a list of objects, and only 841 milliseconds to create 1 million structs
and add them to a list of structs.

So, not only are you able to make structs immutable and thread-safe as they cannot be
modified between threads, but they also perform a lot faster when compared to objects.
Therefore, if you are dealing with large amounts of data, structs can save you a lot of
processing time. Not only that but if you are charged per cycle of execution time by your
cloud computing service, then using structs over objects is going to save you money.

Let's now have a look at writing third-party API tests for the APIs that you will be using.

[ 267 ]
Designing and Developing APIs Chapter 9

Testing third-party APIs

You may ask "Why should I test third-party APIs?" Well, that is a good question. The reason
why you should test third-party APIs is that just like your own code, third-party code is
susceptible to programming errors. I remember once running into some real difficulty on a
document processing website I was building for a law firm. After much investigation, I
found the problem was down due to faulty JavaScript embedded in the Microsoft API that I
was using. The following screenshot is of the GitHub Issues page for the Microsoft
Cognitive Toolkit, which has 738 outstanding issues:

[ 268 ]
Designing and Developing APIs Chapter 9

As you can see from the Microsoft Cognitive Toolkit, third-party APIs do have issues. That
means as the programmer, the onus is on you to ensure that the third-party APIs that you
employ work as expected. Should you encounter any bugs, then it is good practice to
inform the third party of the bugs. If the API is open source and you have access to the
source code, you can even check out the code and submit your own fixes.

Whenever you encounter bugs in third-party code that will not be addressed in time for
you to meet your deadlines, then one option you have available to you is to write
a wrapper class that has all the same constructors, methods, and properties and makes
them call the same constructors, methods, and properties on the third-party class, with the
exception that you write your own bug-free version of the third-party property or method
that has the bug in it. Chapter 11, Addressing Cross-Cutting Concerns, provides sections on
the proxy pattern and decorator pattern, which will help you write wrapper classes.

Testing your own APIs

In Chapter 6, Unit Testing, and Chapter 7, End-to-End System Testing, you saw, with code
examples, how to test your own code. You should always test your own APIs as it is
important to have complete trust in the quality of your APIs. Therefore, as a programmer,
you should always unit test your code before you pass it on to quality assurance. Quality
assurance should then run integration and regression testing on the API to ensure it meets
the company's agreed level of quality.

Your API may do exactly what the business has asked for and be perfect without bugs; but
when it's integrated with the system, do peculiar things happen that you were unable to
test for in certain situations? Often, I have encountered situations in development teams
where code will work on one person's computer but not on other computers. Yet, there
often seems to be no logical reason for this to be the case. These problems can be incredibly
frustrating and even time-consuming to get to the bottom of. But you want these problems
to be ironed out before you pass on your code for quality assurance, and most definitely
before it is released into production. Having to deal with customer bugs is not always a
pleasant experience.

Testing your programs should involve the following:

When given the correct range of values, the method under test outputs the
correct result.
When given the incorrect range of values, the method provides the appropriate
response without crashing.

[ 269 ]
Designing and Developing APIs Chapter 9

Remember that your API should only include what the business has asked for and should
not make the internal details accessible to clients. This is where the product backlog, which
is part of the Scrum project management methodology, is useful.

The product backlog is the list of new features and technical debts that you and your team
will be working on. Each item in the product backlog will have a description and
acceptance criteria, as shown in the following screenshot:

You write your unit tests around the acceptance criteria. Your tests will include the normal
path of execution and abnormal paths of execution. Using this screenshot as an example,
we have two acceptance criteria:

Data is successfully obtained from third-party APIs.

Data is successfully stored in Cosmos DB.

In these two acceptance criteria, we know we will be calling APIs that obtain data. That
data will be obtained from third parties. Once obtained, the data will then be stored in the
database. On the face of it, this specification that we have to work with is quite vague. In
real life, I have found that this is often the case.

Given the vagueness of the specification, we will make the assumption that the
specification will be generic and will apply to different API calls, and we can assume that
the data returned is JSON data. We will also make the assumption that the returned JSON
data will be stored in its raw form in a Cosmos DB database.

[ 270 ]
Designing and Developing APIs Chapter 9

So, what tests can we write for our first acceptance criteria? Well, we can write the
following test cases:

1. When given an URL with a parameter list, assert that we receive a status of
200 and JSON returned for a GET request when all the correct information is
supplied.
2. Assert that we receive a status of 401 when an unauthorized GET request has
been made.
3. Assert that we receive a status of 403 when the authenticated user is forbidden
from accessing the resource.
4. Assert that we receive a status of 500 when the server is down.

What tests can we write for our second acceptance criteria? Well, we can write the
following test cases:

1. Assert that unauthorized access to the database is denied.

2. Assert that the API handles cases where the database is unavailable gracefully.
3. Assert that authorized access to the database is granted.
4. Assert that JSON insertion into the database succeeds.

So, even from such a vague specification, we have been able to obtain eight test cases.
Between them, all these cases test for a successful roundtrip to the third-party server and
back, and then into the database. They also test for various points at which the process
could fail. If all of these tests pass, we have complete confidence in our code and that it will
pass quality control when it leaves our hands as developers.

In the next section, we will look at how to design APIs using RAML.

API design using RAML

In this section, we will discuss designing an API using RAML. You can gain in-depth
knowledge about all aspects of RAML from the RAML website (https://raml.org/
developers/design-your-api). We are going to learn the basics of RAML by designing a
really simple API using API Workbench in Atom. We'll start with the installation.

The first step is to install the packages.

[ 271 ]
Designing and Developing APIs Chapter 9

Installing Atom and API Workbench by MuleSoft

Let's see how to do that:

1. Start by installing Atom from http://atom.io.

2. Then, click on Install a Package:

[ 272 ]
Designing and Developing APIs Chapter 9

3. Then, search for api-workbench by mulesoft and install it:

4. The installation is successful if you find it listed under Packages | Installed

Packages.

Now that we've installed the packages, let's move onto creating the project.

[ 273 ]
Designing and Developing APIs Chapter 9

Creating the project

Let's see how to go about this:

1. Click File | Add Project Folder.

2. Create a new folder or select an existing one. I will create a new folder
called C:\Development\RAML and open it.
3. Add a new file to your project folder called Shop.raml.
4. Right-click in the file and select Add New | Create New API.
5. Give it any name you want and then click on Ok. You have now just created
your first API design.

If you look at the RAML file, you will see that its contents are in human-readable text. The
API we've just created contains a simple GET command that returns a string that contains
the words "Hello World":
#%RAML 1.0
title: Pet Shop
types:
TestType:
type: object
properties:
id: number
optional?: string
expanded:
type: object
properties:
count: number
/helloWorld:
get:
responses:
200:
body:
application/json:
example: |
{
"message" : "Hello World"
}

This is RAML code. You will see that it is pretty similar to JSON in that the code is simple,
human-readable code that is indented. Delete the file. From the Packages menu, select API
Workbench | Create RAML Project. Fill out the Create RAML Project dialog, as in the
following screenshot:

[ 274 ]
Designing and Developing APIs Chapter 9

The settings in this dialog will produce the following RAML code:
#%RAML 1.0
title: Pet Shop
version: v1
baseUri: /petshop
types:
TestType:
type: object
properties:
id: number
optional?: string
expanded:
type: object
properties:
count: number
/helloWorld:
get:
responses:
200:
body:

[ 275 ]
Designing and Developing APIs Chapter 9

application/json:
example: |
{
"message" : "Hello World"
}

The main difference between the last RAML file and the first one you viewed is the
insertion of the version and baseUri properties. These settings also update
your Project folder's content, as follows:

For a very detailed tutorial on this subject, head on over to http://apiworkbench.com/

docs/. This URL also provides details on how to add resources and methods; fill method
bodies and responses; add sub-resources; add examples and types; create and extract
resource types; add resource type parameters, method parameters, and traits; reuse traits,
resource types, and libraries; add more types and resources; extract libraries; and much
more than we can cover in this chapter.

Now that we have a design that is language implementation-agnostic, how do we generate

our API in C#?

[ 276 ]
Designing and Developing APIs Chapter 9

Generating our C# API from our agnostic RAML

design specification
You will need to have, as a minimum, Visual Studio 2019 Community edition installed.
Then, make sure you close Visual Studio. Also, download and install the Visual
Studio MuleSoftInc.RAMLToolsforNET tool. With these tools installed, we will now
proceed through the steps required to produce the skeleton framework of our previously
specified API. This will be accomplished by adding a RAML/OAS contract and importing
our RAML file:

1. In Visual Studio 2019, create a new .NET Framework console application.

2. Right-click on the project and select Add RAML/OAS Contract. This will open
the following dialog:

[ 277 ]
Designing and Developing APIs Chapter 9

3. Click on Upload, and then select your RAML file. The Import
RAML/OAS dialog will then be presented. Fill the dialog out as shown and then
click on Import:

Your project will now be updated with the required dependencies, and new folders and
files will be added to your console application. You will notice three root folders,
called Contracts, Controllers, and Models. In the Contracts folder, we have our
RAML file and the IV1HelloWorldController interface. It contains one
method: Task<IHttpActionResult> Get(). The v1HelloWorldController class
implements the Iv1HelloWorldController interface. Let's have a look at the
implemented Get() method in the controller class:
/// <summary>
/// /helloWorld

[ 278 ]
Designing and Developing APIs Chapter 9

/// </summary>
/// <returns>HelloWorldGet200</returns>
public async Task<IHttpActionResult> Get()
{
// TODO: implement Get - route: helloWorld/helloWorld
// var result = new HelloWorldGet200();
// return Ok(result);
return Ok();
}

In the preceding code, we can see that the code comments out the instantiation of
the HelloWorldGet200 class and the returned result. The HelloWorldGet200 class is our
model class. We can update our model to whatever data we want it to contain. In our
simple example, we won't bother too much with this; we will just return the "Hello
World!" string. Update the uncommented line to the following:

return Ok("Hello World!");

The Ok() method returns a type of OkNegotiatedContentResult<T>. We will call

this Get() method from our Main() method in the Program class. Update
the Main() method, as shown:
static void Main(string[] args)
{
Task.Run(async () =>
{
var hwc = new v1HelloWorldController();
var response = await hwc.Get() as
OkNegotiatedContentResult<string>;
if (response is OkNegotiatedContentResult<string>)
{
var msg = response.Content;
Console.WriteLine($"Message: {msg}");
}
}).GetAwaiter().GetResult();
Console.ReadKey();
}

As we are running asynchronous code in a static method, we must add the work to the
thread pool queue. We then execute our code and wait for the result. Once the code returns,
we simply wait for a keypress and then exit.

[ 279 ]
Designing and Developing APIs Chapter 9

We have created an MVC API within a console app and executed API calls based on the
RAML file that we imported. This same process works for the ASP.NET and ASP.NET Core
websites. We will now extract RAML from an existing API.

Load the dividend calendar API project from earlier on in this chapter. Then, right-click on
the project and select Extract RAML. Then, once the extraction has finished, run your
project. Change the URL to https://localhost:44325/raml. When you extract RAML,
the code generation process adds a RamlController class to your project, along with a
RAML view. You will see that your API is now documented, as shown in the RAML view:

[ 280 ]
Designing and Developing APIs Chapter 9

By using RAML, you can design an API and then generate the structure and you can
reverse engineer an API. The RAML specification helps you design your API and make
changes by modifying the RAML code. You can view the http://raml.org website for
more information on how to get the most out of the RAML specification if you want to
know more. We'll now have a look at Swagger and how to use it in ASP.NET Core 3+
projects.

Well, we've now reached the end of this chapter. Now, we will summarize what we have
achieved and learned.

Summary
In this chapter, we discussed what an API is. Then, we looked at how we can use API
proxies as contracts between ourselves and the consumers of our APIs. This protects our
APIs from direct access by third parties. Next, we looked at a number of design guidelines
for improving the quality of our APIs.

We then went on to discuss Swagger and saw how to document the Weather API with
Swagger. Testing APIs was then covered, and we saw why it is good to test your code and
any third-party code that you use in your projects. Finally, we looked at designing a
language-agnostic API using RAML and translated it into a working project using C#.

In the next chapter, we will write a project to demonstrate securing keys using Azure Key
Vault and securing our own API using API keys. But before then, let's put your brain to
work to see what you have learned.

Questions
1. What does API stand for?
2. What does REST stand for?
3. What are the six constraints of REST?
4. What does HATEOAS stand for?
5. What is RAML?
6. What is Swagger?
7. What is meant by the term well-defined software boundary?
8. Why should you understand the APIs that you are using?

[ 281 ]
Designing and Developing APIs Chapter 9

9. What performs better—structs or objects?

10. Why should you test third-party APIs?
11. Why should you test your own APIs?
12. How can you determine what tests to write for your code?
13. Name three ways to organize code into well-defined software boundaries.

Further reading
https://weblogs.asp.net/sukumarraju/asp-net-web-api-testing-using-
nunit-framework provides a complete example of using NUnit to test web APIs.
https://raml.org/developers/design-your-api shows you how to design your
API with RAML.
http://apiworkbench.com/docs/ provides documentation on using RAML in
Atom to design your APIs.
https://dotnetcoretutorials.com/2017/10/19/using-swagger-asp-net-core/
is a good introduction to using Swagger.
https://swagger.io/about/ takes you to the Swagger About page.
https://httpstatuses.com/ is a list of HTTP status codes.
https://www.greenbytes.de/tech/webdav/rfc5988.html is the web linking
specification RFC 5988.
https://oauth.net/2/ takes you to the OAuth 2.0 home page.
https://en.wikipedia.org/wiki/Domain-driven_design is the Wikipedia page
for domain-driven design.
https://www.packtpub.com/gb/application-development/hands-domain-
driven-design-net-core provides information on the Hands-On Domain-Driven
Design with .NET Core book.
https://www.packtpub.com/gb/application-development/test-driven-
development-c-and-net-core-mvc-video provides information on Test-Driven
Development with C# and .NET Core and MVC.

[ 282 ]
10
Securing APIs with API Keys
and Azure Key Vault
In this chapter, we are going to see how we can keep secrets in Azure Key Vault. We will
also be looking at how we can use API keys to secure our own keys with authentication and
role-based authorization. To gain first-hand experience with API security, we will build a
fully functional FinTech API.

Our API will extract third-party API data using a private key (kept safe in Azure Key
Vault). We will then secure our API with two API keys; one key will be used internally and
a second key will be used by external users.

The following topics are covered in this chapter:

Accessing the Morningstar API

Storing the Morningstar API in Azure Key Vault
Creating the dividend calendar ASP.NET Core web application in Azure
Publishing our web application
Using an API key to secure our dividend calendar API
Testing our API key's security
Adding the dividend calendar code
Throttling our API
Securing APIs with API Keys and Azure Key Vault Chapter 10

You will understand the basics of good API design and you will be armed with the
knowledge needed to push your API abilities forward. This chapter will assist you in
gaining the following skills:

Securing an API with a client API key

Storing and retrieving secrets using Azure Key Vault
Using Postman to execute API commands that post and get data
Applying for and using third-party APIs on RapidAPI.com
Throttling API usage
Writing FinTech APIs that leverage online financial data

Before we continue, make sure you implement the following technical requirements to get
the most out of this chapter.

Technical requirements
We will be using the following technologies in this chapter to write an API:

Visual Studio 2019 Community edition or higher

Your own personal Morningstar API key from https://rapidapi.com/
integraatio/api/morningstar1
RestSharp (http://restsharp.org/)
Swashbuckle.AspNetCore 5 or higher
Postman (https://www.postman.com/)
Swagger (https://swagger.io)

Undertaking the API project – dividend

calendar
The best way to learn is by doing. So, we will build a working API and secure it. The API
won't be perfect and there will be room for improvement. However, you are free to
implement these improvements yourself and expand on the project as you wish. The main
goal here is to have a fully functioning API that does one thing: return financial data that
lists all the company dividends that will be paid in the current year.

[ 284 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

Our dividend calendar API, which we will be building in this chapter, is

an API that is authenticated with an API key. Depending on what key is
used, authorization will determine whether the user is internal or external.
The controller will then execute the appropriate method, depending on
the type of user. Only the internal user method will be implemented, but
you are free to implement the external user method yourself as a training
exercise.

The internal method extracts an API key from Azure Key Vault and
executes various API calls to a third-party API. The data is returned in
JavaScript Object Notation (JSON) format, deserialized into objects and
then processed to extract future dividend payments, which are added to a
list of dividends. This list is then returned to the caller in JSON format.
The end result is a JSON file that has all the scheduled dividend payments
for the current year. The end user can then take this data and convert it
into a list of dividends that can be queried using LINQ.

The project we will be building in this chapter is a web API that returns processed
JSON from third-party financial APIs. Our project will obtain a list of companies from a
given stock exchange. We will then loop through these companies to obtain their dividend
data. The dividend data will then be processed for the current year. So, what we will end
up returning to the API caller is JSON data. This JSON data will contain a list of companies
and their dividend payment forecast for the current year. The JSON data can then be
converted by the end user into C# objects, and LINQ queries can be performed on these
objects. Queries can be carried out to get x-dividend payments for the next month or
payments due this month, for example.

The APIs that we will be using will be part of the Morningstar API, which is available via
RapidAPI.com. You can sign up for a free Morningstar API key. We will secure our API
with a login system, where users log in using an email address and password. You will also
need Postman, as we will be using it to fire the API's POST and GET requests to the dividend
calendar API.

Our solution will contain a single project, which will be an ASP.NET Core application that
targets .NET Framework Core 3.1 or higher. We will now discuss how to access the
Morningstar API.

[ 285 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

Accessing the Morningstar API

Go to https://rapidapi.com/integraatio/api/morningstar1 and request an API access
key. The API is a Freemium API. This means you are allowed a certain number of calls for
free for a limited period, after which you need to pay for its usage. Take some time to look
at the API and its documentation. Pay attention to the pricing plans and keep your key a
secret when you receive it.

The APIs that we are interested in are as follows:

GET /companies/list-by-exchange: This API returns a list of countries for

the specified exchange.
GET /dividends: This API gets all the historical and current dividend payment
information for the specified company.

The first part of the API request is the GET HTTP verb, which is used to retrieve a resource.
The second part of the API request is the resource to GET, which in this case is
/companies/list-by-exchange. As we can see in the second bullet point of the
preceding list, we are getting the /dividends resource.

You can test each API in the browser and see the data that is returned. I recommend you do
this before you continue. This will help you get a feel for what we will be working on. The
basic flow we will be using is getting the list of companies that belong to a
specified exchange, then looping through them to obtain the dividend data. If the dividend
data has a future payment date, then the dividend data will be added to the calendar;
otherwise, it will be discarded. No matter how much dividend data exists for a company,
we are only interested in the first record, which is the most current one.

Now that you have your API key (assuming you are following along with these steps), we
will start to build our API.

Storing the Morningstar API key in Azure Key

Vault
We will be using Azure Key Vault and Managed Service Identity (MSI) with an ASP.NET
Core web application. So, before you continue, you will need an Azure subscription. For
new customers, there is a free 12-month offer available at http://azure.microsoft.com/
en-us/free.

[ 286 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

As web developers, it is important not to store secrets in code because code can be reverse-
engineered. If code is open source, then there is the danger of uploading personal or
enterprise keys to a public version control system. A way around this is to store secrets
securely, but this gives rise to a dilemma. To access secret keys, we need to be
authenticated. So, how do we overcome this dilemma?

We can overcome this dilemma by enabling MSI for our Azure service. As a result, a service
principal is produced by Azure. This service principal is used by applications developed by
the user to access resources on Microsoft Azure. For the service principal, you can use a
certificate or a username and password, along with any role you choose that has the
required set of permissions.

The person who controls the Azure account is in control of what specific tasks each service
can perform. It is often best to start with full restrictions and only add capabilities as and
when they are needed. The following diagram shows the relationships between our
ASP.NET Core web applications, MSI, and our Azure service:

[ 287 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

Azure Active Directory (Azure AD) is employed by MSI to inject the service principal for
the service instance. An Azure resource known as a local metadata service is used to obtain
an access token and will be used to authenticate service access to the Azure key vault.

The code then calls a local metadata service that is available on the Azure resource to get
the access token. The access token extracted from the local MSI endpoint is then used by
our code to authenticate to an Azure Key Vault service.

Open the Azure CLI and type az login to log in to Azure. Once we are logged in, we can
create a resource group. Azure resource groups are logical containers into which Azure
resources are deployed and managed. The following command creates a resource group in
the East US location:
az group create --name "<YourResourceGroupName>" --location "East US"

Use this resource group throughout the rest of the chapter. We will now move on to
creating our key vault. The creation of a key vault requires the following information:

The name of the key vault, which is a string that is between 3 to 24 characters
long and can only contain the 0-9, a-z, A-Z, and - (hyphen) characters
The name of the resource group
The location—for example, East US or West US

In the Azure CLI, enter the following command:

az keyvault create --name "<YourKeyVaultName>" --resource-group
"<YourResourceGroupName> --location "East US"

Only your Azure account is authorized to perform operations on your new vault at this
stage. You can add other accounts if necessary.

The main key that we need to add to our project is MorningstarApiKey. To add the
Morningstar API key to your key vault, type the following command:
az keyvault secret set --vault-name "<YourKeyVaultName>" --name
"MorningstarApiKey" --value "<YourMorningstarApiKey>"

[ 288 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

Your key vault now stores your Morningstar API key. To check that the value is stored
correctly, type the following command:
az keyvault secret show --name "MorningstarApiKey" --vault-name
"<YourKeyVaultName>"

You should now see your secret displayed in the console window, which shows the key
and value for the stored secret.

Creating the dividend calendar ASP.NET

Core web application in Azure
To complete this stage of the project, you will need Visual Studio 2019 with the ASP.NET
and web development workload installed:

1. Create a new ASP.NET Core web application:

[ 289 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

2. Make sure API is selected with No Authentication set:

3. Click on Create to scaffold your new project. Then, run your project. By default,
an example weather forecast API is defined, and it outputs the following JSON
code in the browser window:
[{"date":"2020-04-13T20:02:22.8144942+01:00","temperatureC":0,"temperat
ureF":32,"summary":"Balmy"},{"date":"2020-04-14T20:02:22.8234349+01:00"
,"temperatureC":13,"temperatureF":55,"summary":"Warm"},{"date":"2020-04
-15T20:02:22.8234571+01:00","temperatureC":3,"temperatureF":37,"summary
":"Scorching"},{"date":"2020-04-16T20:02:22.8234587+01:00","temperature
C":-2,"temperatureF":29,"summary":"Sweltering"},{"date":"2020-04-17T20:
02:22.8234602+01:00","temperatureC":-13,"temperatureF":9,"summary":"Coo
l"}]

Next, we will publish our application to Azure.

[ 290 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

Publishing our web application

Before we can publish our web applications, we will first create a new Azure app service to
publish our application to. We will need a resource group to contain our Azure app service,
as well as a new hosting plan that specifies the location, size, and features of the web server
farm that hosts our application. So, let's take care of these requirements, as follows:

1. Make sure you are signed in to your Azure account from Visual Studio. To create
the app service, right-click on the project that you just created and
select Publish from the menu. This will display the Pick a publish target dialog,
as shown:

[ 291 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

2. Select App Service | Create New and click on Create Profile. Create a new
hosting plan, as in the following example:

[ 292 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

3. Then, make sure you provide a name, select a subscription, and select your
resource group. It is recommended that you also set the Application Insights
setting:

[ 293 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

4. Click on Create to create your app service. Once it's created, your Publish screen
should look like this:

5. At this stage, you can click on the site URL. This will load your site URL in the
browser. If your service is successfully configured and running, your browser
should display the following page:

[ 294 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

6. Let's publish our API. Click on the Publish button. When the web page runs, it
will display an error page. Modify the URL
to https://dividend-calendar.azurewebsites.net/weatherforecast.
The web page should now display the weather forecast API JSON code:
[{"date":"2020-04-13T19:36:26.9794202+00:00","temperatureC":40,"tem
peratureF":103,"summary":"Hot"},{"date":"2020-04-14T19:36:26.979734
6+00:00","temperatureC":7,"temperatureF":44,"summary":"Bracing"},{"
date":"2020-04-15T19:36:26.9797374+00:00","temperatureC":8,"tempera
tureF":46,"summary":"Scorching"},{"date":"2020-04-16T19:36:26.97973
89+00:00","temperatureC":11,"temperatureF":51,"summary":"Freezing"}
,{"date":"2020-04-17T19:36:26.9797403+00:00","temperatureC":3,"temp
eratureF":37,"summary":"Hot"}]

Our service is now live. If you log in to your Azure portal and visit the resource group for
your hosting plan, you will see four resources. These resources are as follows:

App Service: dividend-calendar

Application Insights: dividend-calendar
App Service plan: DividendCalendarHostingPlan

[ 295 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

Key vault: Whatever your key vault is called. In my case, it's Keys-APIs, as
shown here:

If you click on your app service from the Azure portal home page (https://portal.azure.
com/#home), you will see that you can browse to your service, as well as stop, restart, and
delete your app service:

[ 296 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

Now that we have our project in place with Application Insights and our Morningstar API
key is stored securely, we can start building our dividend calendar.

Using an API key to secure our dividend

calendar API
To secure access to our dividend calendar API, we are going to use a client API key. There
are many ways to share client keys with your clients, but we will not be discussing them
here. You can come up with your own strategies. What we will be focusing on is how to
enable authenticated and authorized client access to our API.

To keep things simple, we will be using the repository pattern. The repository pattern
helps to decouple our program from the underlying data store. This pattern improves
maintainability and allows you to change the underlying data store without affecting the
program. For our repository, our keys will be defined in a class, but in a commercial
project, you would store the keys in a data store, such as Cosmos DB, SQL Server, or Azure
Key Vault. You decide the strategy that best suits your needs, which is the main reason
why we use the repository pattern as you are in control of the underlying data source for
your own needs.

Setting up the repository

We are going to start by setting up our repository:

1. Add a new folder to your project called Repository. Then, add a new interface
called IRepository and a class that will implement IRepository, called
InMemoryRepository. Modify your interface, as follows:

using CH09_DividendCalendar.Security.Authentication;
using System.Threading.Tasks;

namespace CH09_DividendCalendar.Repository
{
public interface IRepository
{
Task<ApiKey> GetApiKey(string providedApiKey);
}
}

[ 297 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

2. This interface defines one method for retrieving the API key. We have not yet
defined the ApiKey class and we will do that later. Now, let's
implement InMemoryRepository. Add the following using statements:
using CH09_DividendCalendar.Security.Authentication;
using CH09_DividendCalendar.Security.Authorisation;
using System;
using System.Collections.Generic;
using System.Linq;
using System.Threading.Tasks;

3. The security namespaces will be created when we start adding the

authentication and authorization classes. Modify the Repository class to
implement the IRepository interface. Add the member variable that will hold
our API keys, and then add the GetApiKey() method:
public class InMemoryRepository : IRepository
{
private readonly IDictionary<string, ApiKey> _apiKeys;

public Task<ApiKey> GetApiKey(string providedApiKey)

{
_apiKeys.TryGetValue(providedApiKey, out var key);
return Task.FromResult(key);
}
}

4. The InMemoryRepository class implements the GetApiKey() method

of IRepository. This returns a dictionary of API keys. These keys will be stored
in our _apiKeys dictionary member variable. Now, we'll add our constructor:
public InMemoryRepository()
{
var existingApiKeys = new List<ApiKey>
{
new ApiKey(1, "Internal", "C5BFF7F0-B4DF-475E-A331-
F737424F013C", new DateTime(2019, 01, 01),
new List<string>
{
Roles.Internal
}),
new ApiKey(2, "External", "9218FACE-3EAC-6574-
C3F0-08357FEDABE9", new DateTime(2020, 4, 15),
new List<string>
{
Roles.External

[ 298 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

})
};

_apiKeys = existingApiKeys.ToDictionary(x => x.Key, x => x);

}

5. Our constructor creates a new list of API keys. It creates an internal API key for
internal use only and an external API key for external use only. It then converts
the list into a dictionary and stores the dictionary in _apiKeys. So, we now have
our repository in place.
6. We will be using an HTTP header called X-Api-Key. This will store the client's
API key, which will be passed into our API for authentication and authorization.
Add a new folder to the project called Shared, and then add a new file
called ApiKeyConstants. Update the file with the following code:
namespace CH09_DividendCalendar.Shared
{
public struct ApiKeyConstants
{
public const string HeaderName = "X-Api-Key";
public const string MorningstarApiKeyUrl
=
"https://<YOUR_KEY_VAULT_NAME>.vault.azure.net/secrets/MorningstarA
piKey";
}
}

This file contains two constants—the header name, which will be used when
establishing the user's identity, and the URL for the Morningstar API key, which
is stored in the Azure key vault that we created earlier.

7. Since we will be handling JSON data, we need to set our JSON naming policy.
Add a folder to your project called Json. Then, add a class called
DefaultJsonSerializerOptions:

using System.Text.Json;

namespace CH09_DividendCalendar.Json
{
public static class DefaultJsonSerializerOptions
{
public static JsonSerializerOptions Options => new
JsonSerializerOptions
{
PropertyNamingPolicy = JsonNamingPolicy.CamelCase,
IgnoreNullValues = true

[ 299 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

};
}
}

The DefaultJsonSerializerOptions class sets our JSON naming policy to ignore null
values and to use camel case names.

We will now start adding authentication and authorization to our API.

Setting up authentication and authorization

We will now start work on the security classes for authentication and authorization. It is
good to clarify what we mean by authentication and authorization first. Authentication is
establishing whether the user is authorized to access our API. Authorization is establishing
what permissions the user has once they gain access to our API.

Adding authentication
Before we continue, add a Security folder to your project and then under that folder, add
the Authentication and Authorisation folders. We will start by adding our
Authentication classes; the first class that we will add to our Authentication folder is
ApiKey. Add the following properties to ApiKey:

public int Id { get; }

public string Owner { get; }
public string Key { get; }
public DateTime Created { get; }
public IReadOnlyCollection<string> Roles { get; }

These properties store information that pertains to the specified API key and its owner. The
properties are set via the constructor:
public ApiKey(int id, string owner, string key, DateTime created,
IReadOnlyCollection<string> roles)
{
Id = id;
Owner = owner ?? throw new ArgumentNullException(nameof(owner));
Key = key ?? throw new ArgumentNullException(nameof(key));
Created = created;
Roles = roles ?? throw new ArgumentNullException(nameof(roles));
}

[ 300 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

The constructor sets the API key properties. If a person fails authentication, then they will
be notified with an Error 403 Unauthorized message. So, let's now define our
UnauthorizedProblemDetails class:

public class UnauthorizedProblemDetails : ProblemDetails

{
public UnauthorizedProblemDetails(string details = null)
{
Title = "Forbidden";
Detail = details;
Status = 403;
Type = "https://httpstatuses.com/403";
}
}

This class inherits the Microsoft.AspNetCore.Mvc.ProblemDetails class. The

constructor takes a single parameter of the string type, which defaults to null. You can
pass details into this constructor to provide more information if required. Next, we add
AuthenticationBuilderExtensions:

public static class AuthenticationBuilderExtensions

{
public static AuthenticationBuilder AddApiKeySupport(
this AuthenticationBuilder authenticationBuilder,
Action<ApiKeyAuthenticationOptions> options
)
{
return authenticationBuilder
.AddScheme<ApiKeyAuthenticationOptions,
ApiKeyAuthenticationHandler>
(ApiKeyAuthenticationOptions.DefaultScheme, options);
}
}

This extension method adds API key support to the authentication service, which will be
set in the ConfigureServices method of the Startup class. Now, add the
ApiKeyAuthenticationOptions class:

public class ApiKeyAuthenticationOptions : AuthenticationSchemeOptions

{
public const string DefaultScheme = "API Key";
public string Scheme => DefaultScheme;
public string AuthenticationType = DefaultScheme;
}

[ 301 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

The ApiKeyAuthenticationOptions class inherits the AuthenticationSchemeOptions

class. We set the default scheme to use API key authentication. The final part of our
authorization is to build up our ApiKeyAuthenticationHandler class. As the name
suggests, this is the main class for validating the API key, ensuring the client is authorized
to access and use our API:
public class ApiKeyAuthenticationHandler :
AuthenticationHandler<ApiKeyAuthenticationOptions>
{
private const string ProblemDetailsContentType =
"application/problem+json";
private readonly IRepository _repository;
}

Our ApiKeyAuthenticationHandler class inherits from AuthenticationHandler and

uses ApiKeyAuthenticationOptions. We define the content type for the problem details
(exception information) as application/problem+json. We also provide a placeholder
for our API key repository using the _repository member variable. The next step is to
declare our constructor:
public ApiKeyAuthenticationHandler(
IOptionsMonitor<ApiKeyAuthenticationOptions> options,
ILoggerFactory logger,
UrlEncoder encoder,
ISystemClock clock,
IRepository repository
) : base(options, logger, encoder, clock)
{
_repository = repository ?? throw new
ArgumentNullException(nameof(repository));
}

Our constructor passes the ApiKeyAuthenticationOptions, ILoggerFactory,

UrlEncoder, and ISystemClock parameters to the base class. Explicitly, we set the
repository. If the repository is null, we throw a null argument exception with the name of
the repository. Let's add our HandleChallengeAsync() method:
protected override async Task HandleChallengeAsync(AuthenticationProperties
properties)
{
Response.StatusCode = 401;
Response.ContentType = ProblemDetailsContentType;
var problemDetails = new UnauthorizedProblemDetails();
await Response.WriteAsync(JsonSerializer.Serialize(problemDetails,
DefaultJsonSerializerOptions.Options));
}

[ 302 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

The HandleChallengeAsync() method returns an Error 401 Unauthorized response

when the user challenge fails. Now, let's add our HandleForbiddenAsync() method:
protected override async Task HandleForbiddenAsync(AuthenticationProperties
properties)
{
Response.StatusCode = 403;
Response.ContentType = ProblemDetailsContentType;
var problemDetails = new ForbiddenProblemDetails();
await Response.WriteAsync(JsonSerializer.Serialize(problemDetails,
DefaultJsonSerializerOptions.Options));
}

The HandleForbiddenAsync() method returns an Error 403 Forbidden response

when the user permission check fails. Now, we need to add a final method that returns
AuthenticationResult:

protected override async Task<AuthenticateResult> HandleAuthenticateAsync()

{
if (!Request.Headers.TryGetValue(ApiKeyConstants.HeaderName, out var
apiKeyHeaderValues))
return AuthenticateResult.NoResult();
var providedApiKey = apiKeyHeaderValues.FirstOrDefault();
if (apiKeyHeaderValues.Count == 0 ||
string.IsNullOrWhiteSpace(providedApiKey))
return AuthenticateResult.NoResult();
var existingApiKey = await _repository.GetApiKey(providedApiKey);
if (existingApiKey != null) {
var claims = new List<Claim> {new Claim(ClaimTypes.Name,
existingApiKey.Owner)};
claims.AddRange(existingApiKey.Roles.Select(role => new
Claim(ClaimTypes.Role, role)));
var identity = new ClaimsIdentity(claims,
Options.AuthenticationType);
var identities = new List<ClaimsIdentity> { identity };
var principal = new ClaimsPrincipal(identities);
var ticket = new AuthenticationTicket(principal, Options.Scheme);
return AuthenticateResult.Success(ticket);
}
return AuthenticateResult.Fail("Invalid API Key provided.");
}

[ 303 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

The code we've just written checks whether our header exists. If the header is not present,
then AuthenticateResult() returns a Boolean value of true for the None property,
indicating that there was no information provided for this request. We then check whether
the header has a value. If no value is provided, the return value indicates that no
information was provided for this request. We then obtain our server-side key from our
repository using the client-side key.

If the server-side key is null, then a failed AuthenticationResult() instance is returned,

indicating that the provided API key is invalid, as identified in the Failure property of
the Exception type. Otherwise, the user is deemed authentic and is allowed to access our
API. For valid users, we set the claims for their identities and then pass back a successful
AuthenticateResult() instance.

So, we have our authentication sorted. Now, we need to work on our authorization.

Adding authorization
Our authorization classes will be added to the Authorisation folder. Add
the Roles struct with the following code:
public struct Roles
{
public const string Internal = "Internal";
public const string External = "External";
}

We expect our API to be used internally and externally. However, for our minimum viable
product, only the code for internal users will be implemented. Now, add
the Policies struct:
public struct Policies
{
public const string Internal = nameof(Internal);
public const string External = nameof(External);
}

[ 304 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

In our Policies structure, we have added two policies that will be used for internal and
external clients. Now, we'll add the ForbiddenProblemDetails class:
public class ForbiddenProblemDetails : ProblemDetails
{
public ForbiddenProblemDetails(string details = null)
{
Title = "Forbidden";
Detail = details;
Status = 403;
Type = "https://httpstatuses.com/403";
}
}

This class provides the forbidden problem details if one or more permissions are not
available to the authenticated user. You can pass a string into this class's constructor to
provide relevant information if required.

For our authorization, we will need to add authorization requirements and handlers for
both internal and external clients. We'll add the ExternalAuthorisationHandler class
first:
public class ExternalAuthorisationHandler :
AuthorizationHandler<ExternalRequirement>
{
protected override Task HandleRequirementAsync(
AuthorizationHandlerContext context,
ExternalRequirement requirement
)
{
if (context.User.IsInRole(Roles.External))
context.Succeed(requirement);
return Task.CompletedTask;
}

public class ExternalRequirement : IAuthorizationRequirement

{
}

[ 305 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

The ExternalRequirement class is an empty class that implements

the IAuthorizationRequirement interface. Now, add
the InternalAuthorisationHandler class:
public class InternalAuthorisationHandler :
AuthorizationHandler<InternalRequirement>
{
protected override Task HandleRequirementAsync(
AuthorizationHandlerContext context,
InternalRequirement requirement
)
{
if (context.User.IsInRole(Roles.Internal))
context.Succeed(requirement);
return Task.CompletedTask;
}
}

The InternalAuthorisationHandler class handles the authorization of the internal

requirement. If the context user is assigned to the internal role, then permission is granted.
Otherwise, permission is denied. Let's add the required InternalRequirement class:
public class InternalRequirement : IAuthorizationRequirement
{
}

Here, the InternalRequirement class is an empty class that implements

the IAuthorizationRequirement interface.

We now have our authentication and authorization classes in place. So, it is now time to
update our Startup class to wire up the security classes. Start by modifying
the Configure() method:
public void Configure(IApplicationBuilder app, IHostEnvironment env)
{
if (env.IsDevelopment())
{
app.UseDeveloperExceptionPage();
}
app.UseRouting();
app.UseAuthentication();
app.UseAuthorization();
app.UseEndpoints(endpoints =>
{
endpoints.MapControllers();
});
}

[ 306 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

The Configure() method sets the exception page to the developer page if we are in
development. It then requests the app to use routing to match URIs with the actions in our
controllers. The app is then informed that it should use our authentication and
authorization methods. Finally, the application endpoints are mapped from the controllers.

The final method we need to update to complete our API key authentication and
authorization is the ConfigureServices() method. The first thing we need to do is add
our authentication service with API key support:
services.AddAuthentication(options =>
{
options.DefaultAuthenticateScheme =
ApiKeyAuthenticationOptions.DefaultScheme;
options.DefaultChallengeScheme =
ApiKeyAuthenticationOptions.DefaultScheme;
}).AddApiKeySupport(options => { });

Here, we are setting the default authentication scheme. We add AddApiKeySupport()

using our extension key, as defined in our AuthenticationBuilderExtensions class,
which returns; Microsoft.AspNetCore.Authentication.AuthenticationBuilder.
Our default scheme is set to the API key, as configured in our
ApiKeyAuthenticationOptions class. The API key is a constant value that informs the
authentication service that we will be using API key authentication. Now, we need to add
our authorization service:
services.AddAuthorization(options =>
{
options.AddPolicy(Policies.Internal, policy =>
policy.Requirements.Add(new InternalRequirement()));
options.AddPolicy(Policies.External, policy =>
policy.Requirements.Add(new ExternalRequirement()));
});

Here, we are setting our internal and external policies and requirements. These are defined
in our Policies, InternalRequirement, and ExternalRequirement classes.

Well, we've added all of our API key security classes. So, we can now test whether our API
key authentication and authorization is working using Postman.

[ 307 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

Testing our API key security

In this section, we are going to test our API key authentication and authorization using
Postman. Add a class to your Controllers folder called DividendCalendar. Update the
class as follows:
[ApiController]
[Route("api/[controller]")]
public class DividendCalendar : ControllerBase
{
[Authorize(Policy = Policies.Internal)]
[HttpGet("internal")]
public IActionResult GetDividendCalendar()
{
var message = $"Hello from {nameof(GetDividendCalendar)}.";
return new ObjectResult(message);
}

[Authorize(Policy = Policies.External)]
[HttpGet("external")]
public IActionResult External()
{
var message = "External access is currently unavailable.";
return new ObjectResult(message);
}
}

This class will contain all of our dividend calendar API code functionality. Even though
external code will not be used in this initial release of our minimum viable product, we will
be able to test our internal and external authentication and authorization.

1. Open Postman and create a new GET request. For the URL,
use https://localhost:44325/api/dividendcalendar/internal. Click
Send:

[ 308 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

2. As you can see, without the API key present in the API request, we get the
expected 401 Unauthorized status with our forbidden JSON, as defined in
the ForbiddenProblemDetails class. Now, add the x-api-key header with
the C5BFF7F0-B4DF-475E-A331-F737424F013C value. Then, click Send:

[ 309 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

3. You will now have a status of 200 OK. This means that the API request has been
successful. You can see the result of the request in the body. Internal users will
see Hello from GetDividendCalendar. Run the request again, but change the
URL so that the route is external instead of internal. So, the URL should
be https://localhost:44325/api/dividendcalendar/external:

4. You should receive a status of 403 Forbidden with the forbidden JSON. This is
because the API key is a valid API key, but the route is for an external client and
the external client does not have access to the internal API. Change the x-api-
key header value to 9218FACE-3EAC-6574-C3F0-08357FEDABE9. Then, click
Send:

[ 310 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

You will see that you have a status of 200 OK and that the body has the External access
is currently unavailable text.

Good news! Our role-based security system using API key authentication and
authorization is tested and working. So, before we have even added our actual FinTech
API, we have implemented and tested our API key, which is used to secure our FinTech
API. So, we have put the security of our API first before writing a single line of our actual
API. Now, we can start in earnest to build our dividend calendar API functionality,
knowing that it is secure.

Adding the dividend calendar code

Our internal API only has one purpose, which is to build up an array of dividends that are
to be paid out this year. You, however, can build on this project to save the JSON to a file or
database of some type. So, you would only make an internal call once a month to save
money on API calls. However, the external role could access the data from your file or
database as often as needed.

[ 311 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

We already have our controller in place for our dividend calendar API. This security is in
place to prevent unauthenticated and unauthorized users from accessing our
internal GetDividendCalendar() API endpoint. So, all we have to do now is generate the
dividend calendar JSON, which our method will return.

So that you can see what we will be working toward, have a look at the following truncated
JSON response:
[{"Mic":"XLON","Ticker":"ABDP","CompanyName":"AB Dynamics
PLC","DividendYield":0.0,"Amount":0.0279,"ExDividendDate":"2020-01-02T00:00
:00","DeclarationDate":"2019-11-27T00:00:00","RecordDate":"2020-01-03T00:00
:00","PaymentDate":"2020-02-13T00:00:00","DividendType":null,"CurrencyCode"
:null},

...

{"Mic":"XLON","Ticker":"ZYT","CompanyName":"Zytronic
PLC","DividendYield":0.0,"Amount":0.152,"ExDividendDate":"2020-01-09T00:00:
00","DeclarationDate":"2019-12-10T00:00:00","RecordDate":"2020-01-10T00:00:
00","PaymentDate":"2020-02-07T00:00:00","DividendType":null,"CurrencyCode":
null}]

This JSON response is an array of dividends. A dividend consists of the Mic, Ticker,
CompanyName, DividendYield, Amount, ExDividendDate, DeclarationDate,
RecordDate, PaymentDate, DividendType, and CurrencyCode fields. Add a new folder
to your project called Models, and then add the Dividend class with the following code:
public class Dividend
{
public string Mic { get; set; }
public string Ticker { get; set; }
public string CompanyName { get; set; }
public float DividendYield { get; set; }
public float Amount { get; set; }
public DateTime? ExDividendDate { get; set; }
public DateTime? DeclarationDate { get; set; }
public DateTime? RecordDate { get; set; }
public DateTime? PaymentDate { get; set; }
public string DividendType { get; set; }
public string CurrencyCode { get; set; }
}

[ 312 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

Let's see what each of these fields represents:

Mic: ISO 10383 Market Identification Code (MIC), which is where the stock is
listed. See https://www.iso20022.org/10383/iso-10383-market-identifier-
codes for more information.
Ticker: The stock market ticker for the common stock.
CompanyName: The name of the company that owns the stock.
DividendYield: The ratio of the company's annual dividend compared to its
share price. The dividend yield is calculated in terms of percentage and is
calculated with the Dividend Yield = Annual Dividend / Share Price formula.
Amount: The amount that will be paid out to the shareholder per share.
ExDividendDate: The date before which you must purchase the share in order
to receive the next dividend payment.
DeclarationDate: The date that the company declares they are paying a
dividend.
RecordDate: The date that the company looks at its records to determine who
will receive the dividend.
PaymentDate: The date that the shareholders receives the dividend payment.
DividendType: This can be, for example, Cash Dividend, Property
Dividend, Stock Dividend, Scrip Dividend, or Liquidating Dividend.
CurrencyCode: The currency that the amount will be paid in.

The next class we need in our Models folder is the Company class:
public class Company
{
public string MIC { get; set; }
public string Currency { get; set; }
public string Ticker { get; set; }
public string SecurityId { get; set; }
public string CompanyName { get; set; }
}

The Mic and Ticker fields are the same as for our Dividend class.
Between the different API calls, the APIs use different names for the
currency identifier. That is why we have CurrencyCode in Dividend and
Currency in Company. This helps the JSON with the object-mapping
process so that we don't experience formatting exceptions.

[ 313 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

Each of these fields represents the following:

Currency: The currency used to price the stock

SecurityId: The stock market security identifier for the common stock
CompanyName: The name of the company that owns the stock

Our next Models class is called Companies. This class is required to store the companies
that are returned in the initial Morningstar API call. We will be looping through the list of
companies to make further API calls to get each company's record so that we can then make
our API call to get the company's dividend:
public class Companies
{
public int Total { get; set; }
public int Offset { get; set; }
public List<Company> Results { get; set; }
public string ResponseStatus { get; set; }
}

Each of these properties defines the following:

Total: The total number of records returned from the API query
Offset: The record offset
Results: The list of companies returned
ResponseStatus: Provides detailed response information, especially if errors
are returned

Now, we will add the Dividends class. This class holds the list of dividends returned by
the dividends' Morningstar API response:
public class Dividends
{
public int Total { get; set; }
public int Offset { get; set; }
public List<Dictionary<string, string>> Results { get; set; }
public ResponseStatus ResponseStatus { get; set; }
}

Each of these properties is the same as those defined previously, except for
the Results property, which defines a list of dividend payments returned for the specified
company.

[ 314 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

The final class we need to add to our Models folder is the ResponseStatus class. This is
mainly used to store error information:
public class ResponseStatus
{
public string ErrorCode { get; set; }
public string Message { get; set; }
public string StackTrace { get; set; }
public List<Dictionary<string, string>> Errors { get; set; }
public List<Dictionary<string, string>> Meta { get; set; }
}

The properties of this class are as follows:

ErrorCode: The number of the error

Message: The error message
StackTrace: The error diagnostics
Errors: A list of errors
Meta: A list of the error metadata

We now have all the models we need in place. So now, we can start to make our API calls
to build up our dividend payment calendar. In the controller, add a new method
called FormatStringDate(), as follows:
private DateTime? FormatStringDate(string date)
{
return string.IsNullOrEmpty(date) ? (DateTime?)null :
DateTime.Parse(date);
}

This method takes a string date. If the string is null or empty, then null is returned.
Otherwise, the string is parsed and a nullable DateTime value is passed back. We'll also
need a method that extracts our Morningstar API key from our Azure key vault:
private async Task<string> GetMorningstarApiKey()
{
try
{
AzureServiceTokenProvider azureServiceTokenProvider = new
AzureServiceTokenProvider();
KeyVaultClient keyVaultClient = new KeyVaultClient(
new KeyVaultClient.AuthenticationCallback(
azureServiceTokenProvider.KeyVaultTokenCallback
)
);

[ 315 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

var secret = await

keyVaultClient.GetSecretAsync(ApiKeyConstants.MorningstarApiKeyUrl)
.ConfigureAwait(false);
return secret.Value;
}
catch (KeyVaultErrorException keyVaultException)
{
return keyVaultException.Message;
}
}

The GetMorningstarApiKey() method instantiates AzureServiceTokenProvider. It

then creates a new KeyVaultClient object type, which performs cryptographic key
operations. Then, the method awaits the response to get the Morningstar API key from the
Azure key vault. It then passes back the response value. If an error occurs processing a
request, KeyVaultErrorException.Message is returned.

When processing the dividends, we first obtain a list of companies from a stock exchange.
We then loop through these companies and make another call to get the dividends for each
company in that stock exchange. So, we'll start with our method to obtain a list of
companies by MIC. Remember, we are using the RestSharp library. So, if you have not
already installed it, then now is a good time to do so:
private Companies GetCompanies(string mic)
{
var client = new RestClient(
$"https://morningstar1.p.rapidapi.com/companies/list-by-exchange?Mic={mic}"
);
var request = new RestRequest(Method.GET);
request.AddHeader("x-rapidapi-host", "morningstar1.p.rapidapi.com");
request.AddHeader("x-rapidapi-key", GetMorningstarApiKey().Result);
request.AddHeader("accept", "string");
IRestResponse response = client.Execute(request);
return JsonConvert.DeserializeObject<Companies>(response.Content);
}

Our GetCompanies() method creates a new REST client that points to the API URL that
retrieves a list of companies that are listed on the specified stock exchange. The type of the
request is a GET request. We add three headers to our GET request for x-rapidapi-
host, x-rapidapi-key, and accept. Then, we execute the request and return the
deserialized JSON data via the Companies model.

[ 316 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

Now, we will write the methods that return the dividends for the specified exchange and
company. Let's start by adding the GetDividends() method:
private Dividends GetDividends(string mic, string ticker)
{
var client = new RestClient(
$"https://morningstar1.p.rapidapi.com/dividends?Ticker={ticker}&Mic={mic}"
);
var request = new RestRequest(Method.GET);
request.AddHeader("x-rapidapi-host", "morningstar1.p.rapidapi.com");
request.AddHeader("x-rapidapi-key", GetMorningstarApiKey().Result);
request.AddHeader("accept", "string");
IRestResponse response = client.Execute(request);
return JsonConvert.DeserializeObject<Dividends>(response.Content);
}

The GetDividends() method is the same as the GetCompanies() method, except the
request returns the dividends for the specified stock exchange and company. The JSON is
deserialized into an instance of the Dividends object and is returned.

For our final method, we need our minimum viable product to be built into
the BuildDividendCalendar() method. This method is the method that will build up the
dividend calendar JSON that will be returned to the client:
private List<Dividend> BuildDividendCalendar()
{
const string MIC = "XLON";
var thisYearsDividends = new List<Dividend>();
var companies = GetCompanies(MIC);
foreach (var company in companies.Results) {
var dividends = GetDividends(MIC, company.Ticker);
if (dividends.Results == null)
continue;
var currentDividend = dividends.Results.FirstOrDefault();
if (currentDividend == null || currentDividend["payableDt"] ==
null)
continue;
var dateDiff = DateTime.Compare(
DateTime.Parse(currentDividend["payableDt"]),
new DateTime(DateTime.Now.Year - 1, 12, 31)
);
if (dateDiff > 0) {
var payableDate = DateTime.Parse(currentDividend["payableDt"]);
var dividend = new Dividend() {
Mic = MIC,
Ticker = company.Ticker,
CompanyName = company.CompanyName,

[ 317 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

ExDividendDate =
FormatStringDate(currentDividend["exDividendDt"]),
DeclarationDate =
FormatStringDate(currentDividend["declarationDt"]),
RecordDate = FormatStringDate(currentDividend["recordDt"]),
PaymentDate =
FormatStringDate(currentDividend["payableDt"]),
Amount = float.Parse(currentDividend["amount"])
};
thisYearsDividends.Add(dividend);
}
}
return thisYearsDividends;
}

In this version of the API, we hardcode the MIC to "XLON"—the London Stock Exchange.
However, in future releases, this method and the public endpoint could be updated to
accept a MIC as a request parameter. We then add a list variable to hold this year's
dividend payments. Then, we perform our Morningstar API call to extract the list of
companies that are currently listed on the specified MIC. Once the list is returned, we loop
through results. For each company, we then make a further API call to get the complete
dividend record for the MIC and the ticker. If the company has no dividends listed, then we
continue with the next iteration and select the next company.

If the company has dividend records, we get the first record, which will be the latest
dividend payment. We check whether the payable date is null. If the payable date is null,
then we continue on to the next iteration with the next customer. If the payable date is not
st
null, we check whether the payable date is greater than December 31 from the previous
year. If the date difference is greater than 1, then we add a new dividend object to this
year's dividends list. Once we have iterated through all the companies and built up a list of
this year's dividends, we then pass the list back to the calling method.

The final step before we run our project is to update the GetDividendCalendar() method
to call the BuildDividendCalendar() method:
[Authorize(Policy = Policies.Internal)]
[HttpGet("internal")]
public IActionResult GetDividendCalendar()
{
return new
ObjectResult(JsonConvert.SerializeObject(BuildDividendCalendar()));
}

[ 318 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

In the GetDividendCalendar() method, we return a JSON string from the serialized list
of this year's dividends. So, if you run the project in Postman using the internal x-api-
key variable, then you should find that after around 20 minutes, the following JSON is
returned:
[{"Mic":"XLON","Ticker":"ABDP","CompanyName":"AB Dynamics
PLC","DividendYield":0.0,"Amount":0.0279,"ExDividendDate":"2020-01-02T00:00
:00","DeclarationDate":"2019-11-27T00:00:00","RecordDate":"2020-01-03T00:00
:00","PaymentDate":"2020-02-13T00:00:00","DividendType":null,"CurrencyCode"
:null},

...

This query does take a lot of time to run, roughly around 20 minutes, and the results will
change over the course of a year. So, a strategy we could use is to throttle the API to run
once a month and then store the JSON either in a file or a database. Then, this file or
database record is what you would update the external method to call and pass back to
external clients. Let's throttle our API to only run once a month.

Throttling our API

When exposing APIs, you need to throttle them. There are many methods available to do
this, such as limiting the number of simultaneous users or limiting the number of calls
within a given period of time, for example.

In this section, we are going to throttle our API. The method we will use to throttle our API
th
will be to limit our API to run only once a month on the 25 of the month. Add the
following line to your appsettings.json file:
"MorningstarNextRunDate": null,

This value will contain the date that the next API can be executed. Now, add
the AppSettings class at the root of your project, and then add the following property:
public DateTime? MorningstarNextRunDate { get; set; }

[ 319 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

This property will hold the value of the MorningstarNextRunDate key. The next thing to
do is to add our static method, which will be called to add or update an application setting
in the appsetting.json file:
public static void AddOrUpdateAppSetting<T>(string sectionPathKey, T value)
{
try
{
var filePath = Path.Combine(AppContext.BaseDirectory,
"appsettings.json");
string json = File.ReadAllText(filePath);
dynamic jsonObj =
Newtonsoft.Json.JsonConvert.DeserializeObject(json);
SetValueRecursively(sectionPathKey, jsonObj, value);
string output = Newtonsoft.Json.JsonConvert.SerializeObject(
jsonObj,
Newtonsoft.Json.Formatting.Indented
);
File.WriteAllText(filePath, output);
}
catch (Exception ex)
{
Console.WriteLine("Error writing app settings | {0}", ex.Message);
}
}

AddOrUpdateAppSetting() tries to get the file path for the appsettings.json file. It
then reads the JSON from the file. The JSON is then deserialized into a dynamic object. We
then call our method to recursively set the required value. Then, we write the JSON back to
the same file. If an error is encountered, then we output an error message to the console.
Let's write our SetValueRecursively() method:
private static void SetValueRecursively<T>(string sectionPathKey, dynamic
jsonObj, T value)
{
var remainingSections = sectionPathKey.Split(":", 2);
var currentSection = remainingSections[0];
if (remainingSections.Length > 1)
{
var nextSection = remainingSections[1];
SetValueRecursively(nextSection, jsonObj[currentSection], value);
}
else
{
jsonObj[currentSection] = value;
}
}

[ 320 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

The SetValueRecursively() method splits the string at the first apostrophe character. It
then proceeds to recursively process the JSON, moving down the tree. When it gets to
where it needs to be—that is, it finds the required value—the value is then set and the
method returns. Add the ThrottleMonthDay constant to the ApiKeyConstants struct:
public const int ThrottleMonthDay = 25;

This constant is used for our day-of-the-month check when an API request is issued. In
DividendCalendarController, add the ThrottleMessage() method:

private string ThrottleMessage()

{
return "This API call can only be made once on the 25th of each
month.";
}

The ThrottleMessage() method simply returns the message, "This API call can
only be made once on the 25th of each month.". Now, add the following
constructor:
public DividendCalendarController(IOptions<AppSettings> appSettings)
{
_appSettings = appSettings.Value;
}

This constructor provides us with access to the values in the appsettings.json file. Add
these two lines to the end of your Startup.ConfigureServices() method:
var appSettingsSection = Configuration.GetSection("AppSettings");
services.Configure<AppSettings>(appSettingsSection);

These two lines enable the AppSettings class to be dynamically injected into our
controller when we need it. Add the SetMorningstarNextRunDate() method to
the DividendCalendarController class:
private DateTime? SetMorningstarNextRunDate()
{
int month;
if (DateTime.Now.Day < 25)
month = DateTime.Now.Month;
else
month = DateTime.Now.AddMonths(1).Month;
var date = new DateTime(DateTime.Now.Year, month,
ApiKeyConstants.ThrottleMonthDay);
AppSettings.AddOrUpdateAppSetting<DateTime?>(
"MorningstarNextRunDate",
date

[ 321 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

);
return date;
}

The SetMorningstarNextRunDate() method checks whether the current month's day is

less than 25. If the current month's day is less than 25, then the month is set to the current
month so that the API can be run on the 25th of the current month. Otherwise, for days that
are 25 and upward, the month is set to the following month. The new date is then
assembled and the MorningstarNextRunDate key of appsettings.json is then updated
and the nullable DateTime value is returned:
private bool CanExecuteApiRequest()
{
DateTime? nextRunDate = _appSettings.MorningstarNextRunDate;
if (!nextRunDate.HasValue)
nextRunDate = SetMorningstarNextRunDate();
if (DateTime.Now.Day == ApiKeyConstants.ThrottleMonthDay) {
if (nextRunDate.Value.Month == DateTime.Now.Month) {
SetMorningstarNextRunDate();
return true;
}
else {
return false;
}
}
else {
return false;
}
}

CanExecuteApiRequest() gets the current value of the MorningstarNextRunDate

value from the AppSettings class. If DateTime? does not have a value, then the value is
set and assigned to the nextRunDate local variable. If the current month's day does not
equal ThrottleMonthDay, then we return false. If the current month does not equal the
next run date month, then we return false. Otherwise, we set the next API run date to the
25th of the following month and return true.

Finally, we update our GetDividendCalendar() method, as follows:

[Authorize(Policy = Policies.Internal)]
[HttpGet("internal")]
public IActionResult GetDividendCalendar()
{
if (CanExecuteApiRequest())
return new
ObjectResult(JsonConvert.SerializeObject(BuildDividendCalendar()));

[ 322 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

else
return new ObjectResult(ThrottleMessage());
}

When an internal user calls the API now, their request will be validated to see whether it
can run. If it runs, then the serialized JSON for the dividend calendar is returned.
Otherwise, we return the throttle message.

That concludes our project.

Well, we've completed our project. It is not perfect, and there are improvements and
extensions that we can make. The next step would be to document our API and deploy the
API and documentation. We should also add logging and monitoring.

Logging is useful for storing exception details and for tracking how our API is used.
Monitoring is a way to keep an eye on the health of our API so that we can be alerted if
anything goes wrong. This way, we can be proactive in keeping our API up and running. I
will leave you to extend the API as you desire. It will be a good learning exercise for you.

The next chapter addresses cross-cutting concerns. It will give you an idea
about how to address logging and monitoring using aspects and
attributes.

Let's summarise what we have learned.

Summary
In this chapter, you signed up to a third-party API and received your own key. The API key
is stored in your Azure key vault and kept secure from access by unauthorized clients. You
then moved on to create an ASP.NET Core web application and published it to Azure.
Then, you set about securing the web application by using authentication and role-based
authorization.

The authorization we set up is performed using an API key. You used two API keys in this
project—one for internal use and one for external use. The testing of our API and API key
security was performed using the Postman application. Postman is a very good and useful
tool for testing HTTP requests and responses for the various HTTP verbs.

[ 323 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

You then added the dividend calendar API code and enabled internal and external access
based on API keys. The project itself performed a number of different API calls to build up
a list of companies that are expecting to pay dividends to investors. The project then
serializes the objects into JSON format, which is returned to the client. Finally, the project is
throttled to run once a month.

So, by working through this chapter, you have created a FinTech API that you can run once
a month. This API will provide dividend payment information for the current year. Your
clients can deserialize this data and then perform LINQ queries on it to extract data that
meets their specific requirements.

In the next chapter, we will be using PostSharp to implement Aspect-Oriented

Programming (AOP). With our AOP framework, we will learn how to manage common
functionalities such as exception handling, logging, security, and transactions within our
applications. But before that, let's put your brain to work to see what you have learned.

Questions
1. What URL is a good source for hosting your own APIs and accessing third-party
APIs?
2. What are the two required parts for securing an API?
3. What are claims and why should you use them?
4. What do you use Postman for?
5. Why should you use the repository pattern for your data store?

Further reading
https://docs.microsoft.com/en-us/aspnet/web-api/overview/security/
individual-accounts-in-web-api is Microsoft's in-depth guide to web API
security.
https://docs.microsoft.com/en-us/aspnet/web-forms/overview/older-
versions-security/membership/creating-the-membership-schema-in-sql-
server-vb covers creating the ASP.NET membership database.
https://www.iso20022.org/10383/iso-10383-market-identifier-codes is
about ISO 10383 MIC.

[ 324 ]
Securing APIs with API Keys and Azure Key Vault Chapter 10

https://docs.microsoft.com/en-gb/azure/key-vault/vs-key-vault-add-
connected-service covers adding key vault to your web application by using
Visual Studio Connected Services.
https://aka.ms/installazurecliwindows is about the Azure CLI MSI installer.
https://docs.microsoft.com/en-us/azure/key-vault/service-to-service-
authentication is the Azure service-to-service documentation.
https://azure.microsoft.com/en-gb/free/?WT.mc_id=A261C142F is where you
can sign up for your free 12-month subscription to Azure if you are a new
customer.
https://docs.microsoft.com/en-us/azure/key-vault/basic-concepts looks
at the Azure Key Vault basic concepts.
https://docs.microsoft.com/en-us/azure/app-service/app-service-web-
get-started-dotnet covers creating a .NET Core app in Azure.
https://docs.microsoft.com/en-gb/azure/app-service/overview-hosting-
plans provides an Azure App Service plan overview.
https://docs.microsoft.com/en-us/azure/key-vault/tutorial-net-create-
vault-azure-web-app is a tutorial on using Azure Key Vault with an Azure web
app in .NET.

[ 325 ]
11
Addressing Cross-Cutting
Concerns
There are two types of concerns that you need to have when writing clean code—core
concerns and cross-cutting concerns. Core concerns are the reasons for the software and
why it is being developed. Cross-cutting concerns are the concerns that are not part of the
business requirements and that form the core concerns, but must be addressed in all areas
of the code, as illustrated in the following diagram:

It is the cross-cutting concerns that we will be covering in this chapter by building a

reusable class library that you can modify or extend to your liking. Cross-cutting concerns
include configuration management, logging, auditing, security, validation, exception-
handling, instrumentation, transactions, resource pooling, caching, and threading and
concurrency. We will use the decorator pattern and the PostSharp Aspect Framework to
help us build our reusable library, which is injected at compile time.
Addressing Cross-Cutting Concerns Chapter 11

As you read through this chapter, you will see how attribute programming can result in
using a lot less boilerplate code, as well as code that is smaller, more readable, and easier to
maintain and extend. This leaves only the required business code in your methods with the
boilerplate code

We have discussed many of these ideas already. However, they are

mentioned here again as they are cross-cutting concerns.

In this chapter, we'll be covering the following topics:

The decorator pattern

The proxy pattern
Aspect-Oriented Programming (AOP) with PostSharp
Project – cross-cutting concerns reusable library

By the end of this chapter, you will have the skills to do the following:

Implement the decorator pattern.

Implement the proxy pattern.
Apply AOP using PostSharp.
Build your own reusable AOP library that addresses your cross-cutting concerns.

Technical requirements
To get the most out of this chapter, you will need Visual Studio 2019 and PostSharp
installed. For the code files for this chapter, refer to https://github.com/
PacktPublishing/Clean-Code-in-C-/tree/master/CH11. Let's start by looking at the
decorator pattern.

The decorator pattern

The decorator design pattern is a structural pattern that is used to add new functionality to
an existing object without changing its structure. The original class is wrapped in decorator
class wraps and new behaviors and operations are added to an object at runtime:

[ 327 ]
Addressing Cross-Cutting Concerns Chapter 11

The Component interface and the members it contains are implemented by the
ConcreteComponent class and the Decorator class. ConcreteComponent implements
the Component interface. The Decorator class is an abstract class that implements the
Component interface and contains the reference to a Component instance. The Decorator
class is the base class for components. The ConcreteDecorator class inherits from the
Decorator class and provides a decorator for components.

We are going to write an example that wraps an operation in a try/catch block. Both try
and catch will output a string to the console. Create a new .NET 4.8 console application
named CH10_AddressingCrossCuttingConcerns. Then, add a folder called
DecoratorPattern. Add a new interface called IComponent:

public interface IComponent {

void Operation();
}

To keep things simple, our interface only has a single operation of the void type. Now that
we have our interface in place, we need to add an abstract class that implements the
interface. Add a new abstract class called Decorator that implements
the IComponent interface. Add a member variable to store our IComponent object:
private IComponent _component;

[ 328 ]
Addressing Cross-Cutting Concerns Chapter 11

The _component member variable, which stores the IComponent object, is set via the
constructor, as follows:
public Decorator(IComponent component) {
_component = component;
}

In the preceding code, the constructor sets the component we will be decorating. Next, we
add our interface method:
public virtual void Operation() {
_component.Operation();
}

We've declared the Operation() method as virtual so that it can be overridden in the
derived classes. We'll now create our ConcreteComponent class, which
implements IComponent:
public class ConcreteComponent : IComponent {
public void Operation() {
throw new NotImplementedException();
}
}

As you can see, our class consists of one operation, which throws
NotImplementedException. Now, we can write about the ConcreteDecorator class:

public class ConcreteDecorator : Decorator {

public ConcreteDecorator(IComponent component) : base(component) { }
}

The ConcreteDecorator class inherits the Decorator class. The constructor takes
an IComponent parameter and passes it to the base constructor, where the member
variable is then set. Next, we'll override the Operation() method:
public override void Operation() {
try {
Console.WriteLine("Operation: try block.");
base.Operation();
} catch(Exception ex) {
Console.WriteLine("Operation: catch block.");
Console.WriteLine(ex.Message);
}
}

[ 329 ]
Addressing Cross-Cutting Concerns Chapter 11

In our overridden method, we have a try/catch block. In the try block, we write a
message to the console and execute the base class' Operation() method. In
the catch block, when an exception is encountered, a message is written, followed by the
error message. Before we can use our code, we need to update the Program class. Add
the DecoratorPatternExample() method to the Program class:
private static void DecoratorPatternExample() {
var concreteComponent = new ConcreteComponent();
var concreteDecorator = new ConcreteDecorator(concreteComponent);
concreteDecorator.Operation();
}

In our DecoratorPatternExample() method, we create a new concrete component. We

then pass it into the constructor of a new concrete decorator. Then, we call
the Operation() method on the concrete decorator. Add the following two lines to
the Main() method:
DecoratorPatternExample();
Console.ReadKey();

Those two lines execute our example and then wait for the user to press a key before
exiting. Run the code and you should see the same output as in the following screenshot:

That concludes our look at the decorator pattern. Now, it's time to look at the proxy pattern.

[ 330 ]
Addressing Cross-Cutting Concerns Chapter 11

The proxy pattern

The proxy pattern is a structural design pattern providing objects that act as substitutes for
real service objects used by clients. Proxies receive client requests, perform the required
work, and then pass the request to service objects. Proxy objects are interchangeable with
services as they share the same interfaces:

An example of when you would want to use the proxy pattern is when you have a class
that you do not want to change, but where you do need additional behaviors to be added.
Proxies delegate work to other objects. Unless a proxy is a derivative of a service, proxy
methods should finally refer to a Service object.

We will look at a very simple implementation of the proxy pattern. Add a folder to the root
of your Chapter 11 project called ProxyPattern. Add an interface called IService with
a single method to handle a request:
public interface IService {
void Request();
}

[ 331 ]
Addressing Cross-Cutting Concerns Chapter 11

The Request() method performs the work that carries out the request. Both the proxy and
the service will implement this interface to use the Request() method. Now, add
the Service class and implement the IService interface:
public class Service : IService {
public void Request() {
Console.WriteLine("Service: Request();");
}
}

Our Service class implements the IService interface and handles the actual
service Request() method. This Request() method will be called by the Proxy class. The
final step to implementing the proxy pattern is to write the Proxy class:
public class Proxy : IService {
private IService _service;

public Proxy(IService service) {

_service = service;
}

public void Request() {

Console.WriteLine("Proxy: Request();");
_service.Request();
}
}

Our Proxy class implements IService and has a constructor that accepts a
single IService parameter. The Request() method of the Proxy class is called by the
client. The Proxy.Request() method will do what it needs to do and will be responsible
for calling _service.Request(). So that we can see this in action, let's update
our Program class. Add the ProxyPatternExample() call to the Main() method. Then,
add the ProxyPatternExample() method:
private static void ProxyPatternExample() {
Console.WriteLine("### Calling the Service directly. ###");
var service = new Service();
service.Request();
Console.WriteLine("## Calling the Service via a Proxy. ###");
new Proxy(service).Request();
}

[ 332 ]
Addressing Cross-Cutting Concerns Chapter 11

Our test method runs the Request() method of the Service class direction. Then, it runs
the same method via the Request() method of the Proxy class. Run the project and you
should see the following:

Now that you have a working understanding of the decorator and proxy patterns, let's take
a look at AOP with PostSharp.

AOP with PostSharp

AOP can be used with OOP. An aspect is an attribute applied to classes, methods,
parameters, and properties that, at compile-time, weaves code into the class, method,
parameter, or property to which it is applied. This approach allows the cross-cutting
concerns of a program to be moved from the business source code to a class library. The
concerns are added where needed as attributes. The compiler then weaves the required
code in at runtime. This keeps your business code small and readable. In this chapter, we
will be using PostSharp. You can download it from https://www.postsharp.net/download.

So, how does AOP work with PostSharp?

You add the PostSharp package to your project. Then, you annotate your code with
attributes. The C# compiler builds your code into binary, and then PostSharp analyzes the
binary and injects the implementation of the aspects. Although the binaries are modified
with injected code at compile-time, your project's source code remains unaltered. This
means you can keep your code nice, clean, and simple, which in turn makes maintenance,
reuse, and extending existing code bases much easier in the long term.

PostSharp has some really good ready-made patterns for you to utilize. These cover Model-
View-ViewModel (MVVM), caching, multi-threading, longing and architecture validation,
and more. But the good news is that if there is nothing that meets your requirements, then
you can automate your own patterns by extending the aspect framework and/or the
architecture framework.

[ 333 ]
Addressing Cross-Cutting Concerns Chapter 11

With the aspect framework, you develop your simple or composite aspect, apply it to the
code, and validate its usage. As for the architectural framework, you develop your custom
architectural constraints. Before we delve into the cross-cutting concerns, let's briefly take a
look at extending the aspect and architectural frameworks.

You need to add the PostSharp.Redist NuGet package when writing

aspects and attributes. Once done, if you find that your attributes and
aspects are not working, then right-click on the project and select Add
PostSharp to Project. After you've done this, your aspects should work.

Extending the aspect framework

In this section, we are going to develop a simple aspect and apply it to some code. Then, we
will validate the usage of our aspect.

Developing our aspect

Our aspect will be a simple one that is composed of a single transformation. We will derive
our aspect from a primitive aspect class. Then, we will override some methods known as
advice. If you would like to know how to create a composite aspect, you can read how to
do so at https://doc.postsharp.net/complex-aspects.

Injecting behaviors before and after the method execution

The OnMethodBoundaryAspect aspect implements the decorator pattern. You have
already seen how to implement the decorator pattern earlier in this chapter. With this
aspect, you can execute logic before and after the execution of a target method. The
following table provides a list of the advice methods that are available in the
OnMethodBoundaryAspect class:

Advice Description
Used when the method's execution starts, before any
OnEntry(MethodExecutionArgs)
user code.

OnSuccess(MethodExecutionArgs) Used when the method's execution succeeds (that is,

returns without an exception), after any user code.
Used when the method execution fails with an exception,
OnException(MethodExecutionArgs)
after any user code. It is equivalent to a catch block.

[ 334 ]
Addressing Cross-Cutting Concerns Chapter 11

Used when the method execution exits, whether

successfully or with an exception. This advice runs after
any user code and after the
OnExit(MethodExecutionArgs) OnSuccess(MethodExecutionArgs) or
OnException(MethodExecutionArgs) method
of the current aspect. It is equivalent to a finally
block.

For our simple aspect, we are going to look at all the methods in use. Before we begin, add
PostSharp to your project. If you have already downloaded PostSharp, you can right-click
on your project and then select Add PostSharp to Project. After that, add a new folder to
your project called Aspects, and then add a new class called LoggingAspect:
[PSerializable]
public class LoggingAspect : OnMethodBoundaryAspect { }

The [PSerializeable] attribute is a custom attribute that, when applied to a type, causes
PostSharp to generate a serializer for use by PortableFormatter. Now, override the
OnEntry() method:

public override void OnEntry(MethodExecutionArgs args) {

Console.WriteLine("The {0} method has been entered.",
args.Method.Name);
}

The OnEntry() method is executed before any user code. Now, override the OnSuccess()
method:
public override void OnSuccess(MethodExecutionArgs args) {
Console.WriteLine("The {0} method executed successfully.",
args.Method.Name);
}

The OnSuccess() method runs after the user code has completed without exception.
Override the OnExit() method:
public override void OnExit(MethodExecutionArgs args) {
Console.WriteLine("The {0} method has exited.", args.Method.Name);
}

[ 335 ]
Addressing Cross-Cutting Concerns Chapter 11

The OnExit() method executes when the user method completes successfully or
unsuccessfully and exits. It is equivalent to a finally block. Finally, override the
OnException() method:

public override void OnException(MethodExecutionArgs args) {

Console.WriteLine("An exception was thrown in {0}.", args.Method.Name);
}

The OnException() method executes when method execution fails with an exception,
after any user code. It is equivalent to a catch block.

The next step is to write two methods that we can apply LoggingAspect to. We'll add
SuccessfulMethod():

[LoggingAspect]
private static void SuccessfulMethod() {
Console.WriteLine("Hello World, I am a success!");
}

SuccessfulMethod() uses LoggingAspect and prints a message to the console. Now,

let's add FailedMethod():
[LoggingAspect]
private static void FailedMethod() {
Console.WriteLine("Hello World, I am a failure!");
var x = 1;
var y = 0;
var z = x / y;
}

FailedMethod() uses LoggingAspect and prints a message to the console. Then, it

performs a division by zero operations, which results in DivideByZeroException. Call
both of these methods from your Main() method, and then run through your project. You
should see the following output:

[ 336 ]
Addressing Cross-Cutting Concerns Chapter 11

At this point, the debugger will cause the program to exit. That's it. As you can see, creating
your own PostSharp aspects to meet your needs is a simple process. Now, we will look at
adding our own architectural constraint.

Extending the architectural framework

An architectural constraint is the adoption of custom design patterns that must be respected
across all modules. We will implement a scalar constraint that validates an element of code.

Our scalar constraint, called BusinessRulePatternValidation, will validate that any

class deriving from the BusinessRule class must have a nested class named Factory.
Start by adding the BusinessRulePatternValidation class:
[MulticastAttributeUsage(MulticastTargets.Class, Inheritance =
MulticastInheritance.Strict)]
public class BusinessRulePatternValidation : ScalarConstraint { }

MulticastAttributeUsage designates that this validation aspect will only work with
classes and inheritance allowed. Let's override the ValidateCode() method:
public override void CodeValidation(object target) {
var targetType = (Type)target;
if (targetType.GetNestedType("Factory") == null) {
Message.Write(
targetType, SeverityType.Warning,
"10",
"You must include a 'Factory' as a nested type for {0}.",
targetType.DeclaringType,
targetType.Name);
}
}

Our ValidateCode() method checks whether the target object has a nested Factory type.
If the Factory type is not present, then an exception message is written to the output
window. Add the BusinessRule class:
[BusinessRulePatternValidation]
public class BusinessRule { }

[ 337 ]
Addressing Cross-Cutting Concerns Chapter 11

The BusinessRule class is empty and devoid of Factory. It has our

BusinessRulePatternValidation attribute assigned to it, which is an architectural
constraint. Build your project and you will see the message in the output window. We will
now start to build a reusable class library that you can extend and use in your own projects
to address cross-cutting concerns using AOP and the decorator pattern.

Project – cross-cutting concerns reusable

library
In this section, we will be working through writing a reusable library for addressing
various cross-cutting concerns. It will have limited functionality, but it will give you the
knowledge you need to further expand the project for your own needs. The class library
you will be creating will be a .NET standard library so that it can be used for apps that
target both .NET Framework and .NET Core. You will also create a .NET Framework
console application to see the library in action.

Start by creating a new .NET standard class library called CrossCuttingConcerns. Then,
add a .NET Framework console application to the solution called TestHarness. We will be
adding reusable functionality to address various concerns, starting with caching.

Adding the caching concern

Caching is a storage technique for improving performance when accessing various kinds of
resources. The cache used can be memory, a filesystem, or a database. The type of cache
you use will be dependent on the needs of the project. For our demonstration, we will be
using memory caching to keep things simple.

Add a folder called Caching to the CrossCuttingConcerns project. Then, add a class
called MemoryCache. Add the following NuGet packages to the project:

PostSharp
PostSharp.Patterns.Common
PostSharp.Patterns.Diagnostics
System.Runtime.Caching

[ 338 ]
Addressing Cross-Cutting Concerns Chapter 11

Update the MemoryCache class with the following code:

public static class MemoryCache {
public static T GetItem<T>(string itemName, TimeSpan timeInCache,
Func<T> itemCacheFunction) {
var cache = System.Runtime.Caching.MemoryCache.Default;
var cachedItem = (T) cache[itemName];
if (cachedItem != null) return cachedItem;
var policy = new CacheItemPolicy {AbsoluteExpiration =
DateTimeOffset.Now.Add(timeInCache)};
cachedItem = itemCacheFunction();
cache.Set(itemName, cachedItem, policy);
return cachedItem;
}
}

The GetItem() method takes the name of the cached item, itemName, the length of time
the item is to remain in the cache, timeInCache, and the function to call to place the item
in the cache if it is not already there, itemCacheFunction. Add a new class to the
TestHarness project and call it TestClass. Then, add the GetCachedItem() and
GetMessage() methods, as shown:

public string GetCachedItem() {

return MemoryCache.GetItem<string>("Message", TimeSpan.FromSeconds(30),
GetMessage);
}

private string GetMessage() {

return "Hello, world of cache!";
}

The GetCachedItem() method gets a string called "Message" from the cache. If it is not in
the cache, then it will be stored in the cache by the GetMessage() method for 30 seconds.

Update your Main() method in the Program class to call the GetCachedItem() method,
as shown:
var harness = new TestClass();
Console.WriteLine(harness.GetCachedItem());
Console.WriteLine(harness.GetCachedItem());
Thread.Sleep(TimeSpan.FromSeconds(1));
Console.WriteLine(harness.GetCachedItem());

The first call to GetCachedItem() stores the item in the cache and then returns it. The
second call obtains the item from the cache and returns it. The sleeping thread invalidates
the cache, and so the last call stores the item in the cache before returning it.

[ 339 ]
Addressing Cross-Cutting Concerns Chapter 11

Adding file logging capabilities

In our project, the logging, auditing, and instrumentation processes will send their output
to a text file. So, we will need a class to manage adding the files if they don't exist, and then
adding the output to those files and saving them. Add a folder to the class library called
FileSystem. Then, add a class called LogFile. Set the class as public static and add
the following member variables:
private static string _location = string.Empty;
private static string _filename = string.Empty;
private static string _file = string.Empty;

The _location variable is assigned the folder for the entry assembly. The _filename
variable is assigned the name of the file with the file extension. We need to add the Logs
folder at runtime (if it does not exist). So, we will add the AddDirectory() method to the
FileSystem class:

private static void AddDirectory() {

if (!Directory.Exists(_location))
Directory.CreateDirectory("Logs");
}

The AddDirectory() method checks whether the location exists. If it does not exist, then
the directory is created. Next, we need to deal with adding the file if it does not exist. So,
add the AddFile() method:
private static void AddFile() {
_file = Path.Combine(_location, _filename);
if (File.Exists(_file)) return;
using (File.Create($"Logs\\{_filename}")) {

}
}

In the AddFile() method, we combine the location and filename. If the filename already
exists, then we exit the method; otherwise, we create the file. If we don't use the using
statement, we will encounter IOException when we create our first record, but
subsequent saves will be fine. So, by using the using statement, we avoid the exception
and log the data. We can now write a method that actually saves the data to a file. Add the
AppendTextToFile() method:

public static void AppendTextToFile(string filename, string text) {

_location =
$"{Path.GetDirectoryName(Assembly.GetEntryAssembly()?.Location)}\\Logs";
_filename = filename;

[ 340 ]
Addressing Cross-Cutting Concerns Chapter 11

AddDirectory();
AddFile();
File.AppendAllText(_file, text);
}

The AppendTextToFile() method takes a filename and text and sets the location to that of
the entry assembly. It then ensures that the file and directory exist. Then, it saves the text to
the specified file. Our file logging capabilities are now taken care of, so now, we can move
on to look at our logging concern.

Adding the logging concern

Most applications need some form of logging. The usual methods of logging are to the
console, filesystem, event logs, and database. In our project, we will only focus on console
and text file logging. Add a folder called Logging to the class library. Then, add a file
called ConsoleLoggingAspect and update it as follows:
[PSerializable]
public class ConsoleLoggingAspect : OnMethodBoundaryAspect { }

The [PSerializable] attribute informs PostSharp to generate a serializer for use by

PortableFormatter. ConsoleLoggingAspect inherits from
OnMethodBoundaryAspect. The OnMethodBoundaryAspect class has methods that we
can override to add code before a method body executes, after a method body executes,
when a method body executes successfully, and when an exception is encountered. We will
override these methods to output a message to the console. This can be a very useful tool
when it comes to debugging to see whether code actually gets called and whether it
successfully completes or encounters an exception. We will start by overriding the
OnEntry() method:

public override void OnEntry(MethodExecutionArgs args) {

Console.WriteLine($"Method: {args.Method.Name}, OnEntry().");
}

[ 341 ]
Addressing Cross-Cutting Concerns Chapter 11

The OnEntry() method executes before the body of our methods do, and our
override prints out the name of the method been executed and its own name. Next, we'll
override the OnExit() method:
public override void OnExit(MethodExecutionArgs args) {
Console.WriteLine($"Method: {args.Method.Name}, OnExit().");
}

The OnExit() method executes after the body of our methods have finished executing, and
our override prints out the name of the method that has been executed and its own name.
Now, we'll add the OnSuccess() method:
public override void OnSuccess(MethodExecutionArgs args) {
Console.WriteLine($"Method: {args.Method.Name}, OnSuccess().");
}

The OnSuccess() method executes after the body of the method it is applied to has
finished and returns without exception. When our override executes, it prints out the name
of the executed method and its own name. The last method we will override is the
OnException() method:

public override void OnException(MethodExecutionArgs args) {

Console.WriteLine($"An exception was thrown in {args.Method.Name}.
{args}");
}

The OnException() method executes when an exception is encountered, and in our

override, we print out the name of the method and the argument's object. To apply the
attribute, use [ConsoleLoggingAspect]. To add a text file logging aspect, add a class
called TextFileLoggingAspect. TextFileLoggingAspect is identical to
ConsoleLoggingAspect, apart from the contents of the overridden methods. The
OnEntry(), OnExit(), and OnSuccess() methods call the
LogFile.AppendTextToFile() method and append the contents to the Log.txt file. The
OnException() method does the same, except it appends the contents to the
Exception.log file. Here is the OnEntry() example:

public override void OnEntry(MethodExecutionArgs args) {

LogFile.AppendTextToFile("Log.txt", $"\nMethod: {args.Method.Name},
OnEntry().");
}

That is our logging taken care of. Now, we'll move on to adding our exceptions concern.

[ 342 ]
Addressing Cross-Cutting Concerns Chapter 11

Adding the exception-handling concern

It is inevitable with software that exceptions will be experienced by users of the software.
So, there needs to be some way to log them. The normal way of logging exceptions is to
store the error in a file on the user's system, such as with Exception.log. That's what we'll
do in this section. We will inherit from the OnExceptionAspect class and write our
exception data to the Exception.log file, which will be located in the Logs folder of our
application. OnExceptionAspect wraps the tagged method in a try/catch block. Add a
new folder to the class library called Exceptions, and then add a file called
ExceptionAspect with the following code:

[PSerializable]
public class ExceptionAspect : OnExceptionAspect {
public string Message { get; set; }
public Type ExceptionType { get; set; }
public FlowBehavior Behavior { get; set; }

public override void OnException(MethodExecutionArgs args) {

public override Type GetExceptionType(System.Reflection.MethodBase

targetMethod) {
return ExceptionType;
}
}

The ExceptionAspect class is assigned the [PSerializable] aspect and inherits from
OnExceptionAspect. We have three properties: message, ExceptionType, and
FlowBehavior. message contains the exception message, ExceptionType contains the
type of exception encountered, and FlowBehavior determines whether execution
continues once the exception is handled or whether the process terminates. The
GetExceptionType() method returns the type of exception that was thrown. The
OnException() method starts by constructing the error message. It then logs the exception
to file by calling LogFile.AppendTextToFile(). Finally, the flow of the exception's
behavior is set to continue.

[ 343 ]
Addressing Cross-Cutting Concerns Chapter 11

All you have to do to use the [ExceptionAspect] aspect is add it as an attribute to your
method. We have now covered exception-handling. So, we'll move on to adding our
security concern.

Adding the security concern

The security needs will be specific to the project being worked on. The most common
concerns are that users are authenticated and authorized to access and use various parts of
the system. In this section, we will use the decorator pattern to implement a secure
component with role-based methods.

Security is a very large subject in itself and beyond the scope of this book.
There are many good APIs out there, such as the various Microsoft APIs.
Refer to https://docs.microsoft.com/en-us/dotnet/standard/
security/ for more information, and for OAuth 2.0, refer to https://
oauth.net/code/dotnet/. We will leave you to select and implement your
own method of security. In this chapter, we simply add our own custom-
defined security using the decorator pattern. You can use this as a base for
implementing any of the aforementioned security methods.

Add a new folder called Security and add an interface to it called ISecureComponent:
public interface ISecureComponent {
void AddData(dynamic data);
int EditData(dynamic data);
int DeleteData(dynamic data);
dynamic GetData(dynamic data);
}

Our secure component interface contains the preceding four methods, which are self-
explanatory. The dynamic keyword means that any type of data can be passed in as a
parameter and that any type of data can be returned from the GetData() method. Next,
we need an abstract class that implements the interface. Add a class called
DecoratorBase, as shown:

public abstract class DecoratorBase : ISecureComponent {

private readonly ISecureComponent _secureComponent;

public DecoratorBase(ISecureComponent secureComponent) {

_secureComponent = secureComponent;
}
}

[ 344 ]
Addressing Cross-Cutting Concerns Chapter 11

The DecoratorBase class implements ISecureComponent. We declare a member variable

of the ISecureComponent type and set it in the default constructor. We need to add the
missing methods of ISecureComponent. Add the AddData() method:
public virtual void AddData(dynamic data) {
_secureComponent.AddData(data);
}

This method will take any type of data and then pass it into the call to the AddData()
method of _secureComponent. Add the missing methods for EditData(),
DeleteData(), and GetData(). Now, add a class called ConcreteSecureComponent,
which implements ISecureComponent. For each method, write a message to the console.
For the DeleteData() and EditData() methods, also return a value of 1. Return "Hi!"
for GetData(). The ConcreteSecureComponent class is the class that executes the secure
work that we are interested in.

We need a way to validate the user and obtain their role. The role will be checked before
executing any methods. So, add the following struct:
public readonly struct Credentials {
public static string Role { get; private set; }

public Credentials(string username, string password) {

switch (username)
{
case "System" when password == "Administrator":
Role = "Administrator";
break;
case "End" when password == "User":
Role = "Restricted";
break;
default:
Role = "Imposter";
break;
}
}
}

To keep things simple, the struct takes a username and password and sets the appropriate
role. Restricted users have fewer privileges than administrators. The final class for our
security concern is the ConcreteDecorator class. Add the class, as follows:
public class ConcreteDecorator : DecoratorBase {
public ConcreteDecorator(ISecureComponent secureComponent) :
base(secureComponent) { }
}

[ 345 ]
Addressing Cross-Cutting Concerns Chapter 11

The ConcreteDecorator class inherits the DecoratorBase class. Our constructor takes a
type of ISecureComponent and passes it to the base class. Add the AddData() method:
public override void AddData(dynamic data) {
if (Credentials.Role.Contains("Administrator") ||
Credentials.Role.Contains("Restricted")) {
base.AddData((object)data);
} else {
throw new UnauthorizedAccessException("Unauthorized");
}
}

AddMethod() checks the user's role against the allowed Administrator and Restricted
roles. If the user is in one of these roles, then the AddData() method is executed in the base
class; otherwise, UnauthorizedAccessException is thrown. The rest of the methods
follow this same pattern. Override the rest of the methods, but make sure the
DeleteData() method can only be executed by administrators.

We will now put our security concerns to work. Add the following line to the top of the
Program class:

private static readonly ConcreteDecorator ConcreteDecorator = new

ConcreteDecorator(
new ConcreteSecureComponent()
);

We are declaring and instantiating a concrete decorator object and passing in the concrete
secure object. This object will be referenced in our data methods. Update the Main()
method, as follows:
private static void Main(string[] _) {
// ReSharper disable once ObjectCreationAsStatement
new Credentials("End", "User");
DoSecureWork();
Console.WriteLine("Press any key to exit.");
Console.ReadKey();
}

[ 346 ]
Addressing Cross-Cutting Concerns Chapter 11

We assign the username and password to the Credentials struct. This causes Role to be
set. We then call the DoWork() method. The DoWork() method will be responsible for
calling the data methods. We then pause for the user to press any key and exit. Add the
DoWork() method:

private static void DoSecureWork() {

AddData();
EditData();
DeleteData();
GetData();
}

The DoSecureWork() method calls each of the data methods that call the data methods on
the concrete decorator. Add the AddData() method:
[ExceptionAspect(consoleOutput: true)]
private static void AddData() {
ConcreteDecorator.AddData("Hello, world!");
}

[ExceptionAspect] is applied to the AddData() method. This will ensure any errors are
logged to the Exceptions.log file. The parameter is set to true, and so the error message
will also be printed in the console window. The method itself calls the AddData() method
on the ConcreteDecorator class. Add the rest of the methods by following the same
procedure. Then, run your code. You should see the following output:

We now have a working role-based object, complete with exception handling. Our next step
is to implement our validation concern.

[ 347 ]
Addressing Cross-Cutting Concerns Chapter 11

Adding the validation concern

All user-entered data should be validated as it could be malicious, incomplete, or in the
wrong format. You need to ensure that your data is clean and cannot cause harm. For our
demonstration concern, we will implement null validation. Start by adding a folder
called Validation to the class library. Then, add a new class called
AllowNullAttribute:

[AttributeUsage(AttributeTargets.Parameter | AttributeTargets.ReturnValue |
AttributeTargets.Property)]
public class AllowNullAttribute : Attribute { }

This attribute allows nulls on parameters, return values, and properties. Now, add the
ValidationFlags enum to a new file of the same name:

[Flags]
public enum ValidationFlags {
Properties = 1,
Methods = 2,
Arguments = 4,
OutValues = 8,
ReturnValues = 16,
NonPublic = 32,
AllPublicArguments = Properties | Methods | Arguments,
AllPublic = AllPublicArguments | OutValues | ReturnValues,
All = AllPublic | NonPublic
}

These flags are used to determine what items an aspect can be applied to. Next, we'll add a
class called ReflectionExtensions:
public static class ReflectionExtensions {
private static bool IsCustomAttributeDefined<T>(this
ICustomAttributeProvider value) where T
: Attribute {
return value.IsDefined(typeof(T), false);
}

public static bool AllowsNull(this ICustomAttributeProvider value) {

return value.IsCustomAttributeDefined<AllowNullAttribute>();
}

public static bool MayNotBeNull(this ParameterInfo arg) {

return !arg.AllowsNull() && !arg.IsOptional &&
!arg.ParameterType.IsValueType;
}
}

[ 348 ]
Addressing Cross-Cutting Concerns Chapter 11

The IsCustomAttributeDefined() method returns true if the attribute type is defined

on this member, and false otherwise. The AllowsNull() method returns true if the
[AllowNull] attribute is already applied, and false if not. The MayNotBeNull() method
checks to see whether nulls are allowed, whether the parameter is optional, and what type
of value the parameter is. A Boolean value is then returned by performing logical AND
operations on these values. It's time to add DisallowNonNullAspect:
[PSerializable]
public class DisallowNonNullAspect : OnMethodBoundaryAspect {
private int[] _inputArgumentsToValidate;
private int[] _outputArgumentsToValidate;
private string[] _parameterNames;
private bool _validateReturnValue;
private string _memberName;
private bool _isProperty;

public DisallowNonNullAspect() : this(ValidationFlags.AllPublic) { }

public DisallowNonNullAspect(ValidationFlags validationFlags) {

ValidationFlags = validationFlags;
}

public ValidationFlags ValidationFlags { get; set; }

}

This class has the [PSerializable] attribute applied to inform PostSharp to generate a
serializer for PortableFormatter. It also inherits the OnMethodBoundaryAspect class.
We then declare variables to hold the input and output arguments as validated parameter
names, return value validation and the member name, and check whether the item being
validated is a property. The default constructor is configured to allow the validator to be
applied to all public members. We also have a constructor that takes a ValidationFlags
value and a ValidationFlags property. Now, we'll override the
CompileTimeValidate() method:

public override bool CompileTimeValidate(MethodBase method) {

var methodInformation = MethodInformation.GetMethodInformation(method);
var parameters = method.GetParameters();

if (!ValidationFlags.HasFlag(ValidationFlags.NonPublic) &&
!methodInformation.IsPublic) return false;
if (!ValidationFlags.HasFlag(ValidationFlags.Properties) &&
methodInformation.IsProperty)
return false;
if (!ValidationFlags.HasFlag(ValidationFlags.Methods) &&
!methodInformation.IsProperty) return false;

[ 349 ]
Addressing Cross-Cutting Concerns Chapter 11

_parameterNames = parameters.Select(p => p.Name).ToArray();

_memberName = methodInformation.Name;
_isProperty = methodInformation.IsProperty;

var argumentsToValidate = parameters.Where(p =>

p.MayNotBeNull()).ToArray();

_inputArgumentsToValidate =
ValidationFlags.HasFlag(ValidationFlags.Arguments) ?
argumentsToValidate.Where(p => !p.IsOut).Select(p => p.Position).ToArray()
: new int[0];

_outputArgumentsToValidate =
ValidationFlags.HasFlag(ValidationFlags.OutValues) ?
argumentsToValidate.Where(p => p.ParameterType.IsByRef).Select(p =>
p.Position).ToArray() : new int[0];

if (!methodInformation.IsConstructor) {
_validateReturnValue =
ValidationFlags.HasFlag(ValidationFlags.ReturnValues) &&
methodInformation.ReturnParameter.MayNotBeNull();
}

var validationRequired = _validateReturnValue ||

_inputArgumentsToValidate.Length > 0 || _outputArgumentsToValidate.Length >
0;

return validationRequired;
}

This method ensures that the aspect is correctly applied at compile-time. If the aspect is
applied to a wrong type of member, then false is returned. Otherwise, it returns true. We
now override the OnEntry() method:
public override void OnEntry(MethodExecutionArgs args) {
foreach (var argumentPosition in _inputArgumentsToValidate) {
if (args.Arguments[argumentPosition] != null) continue;
var parameterName = _parameterNames[argumentPosition];

if (_isProperty) {
throw new ArgumentNullException(parameterName,
$"Cannot set the value of property '{_memberName}' to
null.");
} else {
throw new ArgumentNullException(parameterName);
}
}
}

[ 350 ]
Addressing Cross-Cutting Concerns Chapter 11

This method checks the input arguments to validate. If any arguments are null, then
ArgumentNullException is thrown; otherwise, the method exits without throwing an
exception. Let's override the OnSuccess() method now:
public override void OnSuccess(MethodExecutionArgs args) {
foreach (var argumentPosition in _outputArgumentsToValidate) {
if (args.Arguments[argumentPosition] != null) continue;
var parameterName = _parameterNames[argumentPosition];
throw new InvalidOperationException($"Out parameter
'{parameterName}' is null.");
}

if (!_validateReturnValue || args.ReturnValue != null) return;

if (_isProperty) {
throw new InvalidOperationException($"Return value of property
'{_memberName}' is null.");
}
throw new InvalidOperationException($"Return value of method
'{_memberName}' is null.");
}

The OnSuccess() method validates the output parameters to validate. If any arguments are
null, then InvalidOperationException will be thrown. The next thing we need to do is
add private class for extracting method information. Add the following class to the
bottom of the DisallowNonNullAspect class before the closing brace:
private class MethodInformation { }

Add the following three constructors to the MethodInformation class:

private MethodInformation(ConstructorInfo constructor) :
this((MethodBase)constructor) {
IsConstructor = true;
Name = constructor.Name;
}

private MethodInformation(MethodInfo method) : this((MethodBase)method) {

IsConstructor = false;
Name = method.Name;
if (method.IsSpecialName &&
(Name.StartsWith("set_", StringComparison.Ordinal) ||
Name.StartsWith("get_", StringComparison.Ordinal))) {
Name = Name.Substring(4);
IsProperty = true;
}
ReturnParameter = method.ReturnParameter;

[ 351 ]
Addressing Cross-Cutting Concerns Chapter 11

private MethodInformation(MethodBase method)

{
IsPublic = method.IsPublic;
}

These constructors differentiate between constructors and methods and perform the
necessary initialization of the method. Add the following method:
private static MethodInformation CreateInstance(MethodInfo method) {
return new MethodInformation(method);
}

The CreateInstance() method creates a new instance of the MethodInformation class

based on the MethodInfo data of the method passed in and returns that instance. Add the
GetMethodInformation() method:

public static MethodInformation GetMethodInformation(MethodBase methodBase)

{
var ctor = methodBase as ConstructorInfo;
if (ctor != null) return new MethodInformation(ctor);
var method = methodBase as MethodInfo;
return method == null ? null : CreateInstance(method);
}

This method casts methodBase to ConstructorInfo and checks for null. If ctor is not
null, then a new MethodInformation class is generated based on the constructor.
However, if ctor is null, then methodBase is cast to MethodInfo. If the method is not
null, then the CreateInstance() method is called, passing in the method. Otherwise,
null is returned. Finally, add the following properties to the class:

public string Name { get; private set; }

public bool IsProperty { get; private set; }
public bool IsPublic { get; private set; }
public bool IsConstructor { get; private set; }
public ParameterInfo ReturnParameter { get; private set; }

These properties are properties of the method that has the aspect applied. We have now
finished writing our validation aspect. You can now use the validator to allow nulls by
attaching the [AllowNull] attribute. You can disallow nulls by attaching
[DisallowNonNullAspect]. Now, we'll add our transaction concern.

[ 352 ]
Addressing Cross-Cutting Concerns Chapter 11

Adding the transaction concern

Transactions are processes that must run to completion or rollback. Add a new folder to the
class library called Transactions, and then add the RequiresTransactionAspect class:
[PSerializable]
[AttributeUsage(AttributeTargets.Method)]
public sealed class RequiresTransactionAspect : OnMethodBoundaryAspect {
public override void OnEntry(MethodExecutionArgs args) {
var transactionScope = new
TransactionScope(TransactionScopeOption.Required);
args.MethodExecutionTag = transactionScope;
}

public override void OnSuccess(MethodExecutionArgs args) {

var transactionScope = (TransactionScope)args.MethodExecutionTag;
transactionScope.Complete();
}

public override void OnExit(MethodExecutionArgs args) {

var transactionScope = (TransactionScope)args.MethodExecutionTag;
transactionScope.Dispose();
}
}

The OnEntry() method starts the transaction, the OnSuccess() method completes the
exception, and the OnExit() method disposes of the transaction. To use the aspect, add
[RequiresTransactionAspect] to your method. To log any exceptions that prevent the
completion of the transaction, you can also assign the
[ExceptionAspect(consoleOutput: false)] aspect. Next, we'll add our resource pool
concern.

Adding the resource pool concern

Resource pools are a good way to improve performance when multiple instances of an
object are expensive to create and destroy. We will create a very simple resource pool for
our needs. Add a folder called ResourcePooling, and then add the ResourcePool class:
public class ResourcePool<T> {
private readonly ConcurrentBag<T> _resources;
private readonly Func<T> _resourceGenerator;

public ResourcePool(Func<T> resourceGenerator) {

_resourceGenerator = resourceGenerator ??

[ 353 ]
Addressing Cross-Cutting Concerns Chapter 11

throw new
ArgumentNullException(nameof(resourceGenerator));
_resources = new ConcurrentBag<T>();
}

public T Get() => _resources.TryTake(out T item) ? item :

_resourceGenerator();
public void Return(T item) => _resources.Add(item);
}

This class creates a new resource generator and stores resources in ConcurrentBag. When
an item is requested, it issues a resource from the pool. If one does not exist, then it is
created, added to the pool, and issued to the caller:
var pool = new ResourcePool<Course>(() => new Course()); // Create a new
pool of Course objects.
var course = pool.Get(); // Get course from pool.
pool.Return(course); // Return the course to the pool.

The code you've just seen shows you how to use the ResourcePool class to create a pool,
obtain a resource, and return it to the pool.

Adding the configuration settings concern

Configuration settings should always be centralized. Since desktop applications store their
settings in the app.config file and web applications store their settings in Web.config,
we can use ConfigurationManager to access the application settings. Add the
System.Configuration.Configuration NuGet library to your class library and test the
harness. Then, add a folder called Configuration and the following Settings class:
public static class Settings {
public static string GetAppSetting(string key) {
return System.Configuration.ConfigurationManager.AppSettings[key];
}

public static void SetAppSettings(this string key, string value) {

System.Configuration.ConfigurationManager.AppSettings[key] = value;
}
}

This class will get and set app settings in the Web.config file and the App.config file. To
include the class in your files, add the following using statement:
using static CrossCuttingConcerns.Configuration.Settings;

[ 354 ]
Addressing Cross-Cutting Concerns Chapter 11

The following code shows you how to use the methods:

Console.WriteLine(GetAppSetting("Greeting"));
"Greeting".SetAppSettings("Goodbye, my friends!");
Console.WriteLine(GetAppSetting("Greeting"));

Using the static import, you don't have to include the class prefix. You can extend the
Settings class to get connection strings or to do whatever configuration you need in your
apps.

Adding the instrumentation concern

Our final cross-cutting concern is that of instrumentation. We use instrumentation to profile
our application and see how long it takes for methods to execute. Add a folder to the class
library called Instrumentation, and then add the InstrumentationAspect class, as
shown:
[PSerializable]
[AttributeUsage(AttributeTargets.Method)]
public class InstrumentationAspect : OnMethodBoundaryAspect {
public override void OnEntry(MethodExecutionArgs args) {
LogFile.AppendTextToFile("Profile.log",
$"\nMethod: {args.Method.Name}, Start Time: {DateTime.Now}");
args.MethodExecutionTag = Stopwatch.StartNew();
}

public override void OnException(MethodExecutionArgs args) {

LogFile.AppendTextToFile("Exception.log",
$"\n{DateTime.Now}: {args.Exception.Source} -
{args.Exception.Message}");
}

public override void OnExit(MethodExecutionArgs args) {

As you can see, the instrumentation aspect only applies to methods, times the start and stop
times of the method, and logs the profile information to the Profile.log file. If an
exception is encountered, then the exception is logged to the Exception.log file.

[ 355 ]
Addressing Cross-Cutting Concerns Chapter 11

We now have a functional and reusable cross-cutting concerns library. Let's summarize
what we have learned in this chapter.

Summary
We've learned some valuable information. We started off by looking at the decorator
pattern and then the proxy pattern. The proxy pattern provides objects that act as
substitutes for real service objects used by clients. A proxy receives a client request,
performs the necessary work, and then passes the request to the service object. Since
proxies share the same interfaces as the services they substitute, they are interchangeable.

After covering the proxy pattern, we then moved onto AOP with PostSharp. We saw how
we can use aspects and attributes together to decorate code so that at compile-time, it
injects code to perform the required operations, such as exception handling, logging,
auditing, and security. We extended the aspect framework by developing our own aspect
and looked at how to use PostSharp and the decorator pattern to address the cross-cutting
concerns of configuration management, logging, auditing, security, validation, exception
handling, instrumentation, transactions, resource pooling, caching, threading, and
concurrency.

In the next chapter, we will look at using tools to help you improve your code quality. But
before then, test your knowledge and then further your reading.

Questions
1. What is a cross-cutting concern and what does AOP stand for?
2. What is an aspect and how do you apply one?
3. What is an attribute and how do you apply one?
4. How do the aspects and attributes work together?
5. How does the build process work with aspects?

Further reading
The PostSharp home page: https://www.postsharp.net/

[ 356 ]
12
Using Tools to Improve Code
Quality
As a programmer, enhancing code quality is one of your chief concerns. Improving the
quality of your code demands the utilization of various tools. Tools designed to improve
your code and also speed up development include code metrics, quick actions,
the JetBrains dotTrace profiler, JetBrains ReSharper, and Telerik JustDecompile.

This is the main thing that we'll be doing in this chapter, with the following topics:

Defining good-quality code

Performing code cleanup and calculating code metrics
Performing code analysis
Using quick actions
Using the JetBrains dotTrace profiler
Using JetBrains ReSharper
Using Telerik JustDecompile

By the end of this chapter, you will have gained the following skills:

Using code metrics to measure software complexity and maintainability

Using quick actions to make changes using a single command
Profiling your code and analyzing bottlenecks with JetBrains dotTrace
Refactoring code using JetBrains ReSharper
Decompiling code and generating a solution using Telerik JustDecompile
Using Tools to Improve Code Quality Chapter 12

Technical requirements
The source code for this book: https://github.com/PacktPublishing/Clean-
Code-in-C-
Visual Studio 2019 Community Edition or higher: https://visualstudio.
microsoft.com/downloads/
Telerik JustDecompile: https://www.telerik.com/products/decompiler.aspx
JetBrains ReSharper Ultimate: https://www.jetbrains.com/resharper/
download/#section=resharper-installer

Defining good-quality code

Good code quality is an essential software property. Financial loss, wasted time and effort,
and even death can result from poor-quality code. High-standard code will have the
qualities of Performance, Availability, Security, Scalability, Maintainability,
Accessibility, Deployability, and Extensibility (PASSMADE).

Performant code is small, only does what it needs to do, and is very fast. Performant code
will not grind a system to a halt. Things that grind a system to a halt are file input/output
(I/O) operations, memory usage, and central processing unit (CPU) usage. Low-
performing code is a candidate for refactoring.

Availability refers to the software being continually available at the required level of
performance. Availability is the ratio between the time the software is functional (tsf) to
the total time it is expected to function (ttef)—for example, tsf=700; ttef =744. 700 / 744 =
0.9409 = 94.09% availability.

Secure code is the code that properly validates input to protect against invalid data formats,
an invalid range data, and malicious attacks and that fully authenticates and authorizes its
users. Secure code is also code that is fault-tolerant. For example, if you are halfway
through transferring money from one account to another and the system crashes, the
operation should ensure the data remains intact, with no money taken from the account in
question.

Scalable code is code that can safely handle exponential growth in the number of users
using the system without the system grinding to a halt. So, whether the software handles
one request per hour or a million requests per hour, there is no degradation in the
performance of the code and no downtime due to excessive load.

[ 358 ]
Using Tools to Improve Code Quality Chapter 12

Maintainability refers to how easy it is to fix bugs and add new functionality. Maintainable
code should be well organized and easy to read. There should be low coupling and high
cohesion so that the code can be easily maintained and extended.

Accessible code is code that people with limited abilities find easy to modify and use
according to their needs. Examples include user interfaces with high contrast, a narrator for
dyslexic and blind people, and so on.

Deployability focuses on the users of the software—will the users be standalone, remote
access, or local network users? Whatever type the user is, the software should be very easy
to deploy without any issues.

Extensibility refers to how easy it is to extend an application by adding new features to it.
Spaghetti code and highly coupled code with low cohesion make this very difficult and
error-prone. Such code can be very hard to read and maintain and is not easy to extend.
Therefore, extensible code is code that is easy to read, easy to maintain, and—thus—easy to
add new features to.

From the PASSMADE requirements of good-quality code, you can easily infer the kinds of
problems that could arise from failing to meet these requirements. Failure to meet these
requirements would lead to poor-performing code that becomes frustrating and unusable.
Clients would be annoyed by increased downtime. Hackers would be able to exploit
vulnerabilities in code that is not secure. The software would degrade exponentially as
more users are added to the system. Code would be hard to fix or extend, and in some cases
impossible to fix or extend. Users with limited abilities would not be able to modify the
software around their limitations, and deployment would be a configuration nightmare.

Code metrics to the rescue. Code metrics enable developers to measure code complexity
and maintainability and thus help us to identify code that is a candidate for refactoring.

With Quick Actions, you can use a single command to refactor C# code, such as extracting
code out into its own method. JetBrains dotTrace allows you to profile your code and find
performance bottlenecks. Further, JetBrains ReSharper is a Visual Studio productivity
extension that enables you to analyze code quality and detect code smells, enforce coding
standards, and refactor code. And Telerik JustDecompile helps you to decompile existing
code for troubleshooting, and to create Intermediate Language (IL), C#, and VB.NET
projects from. This is particularly useful if you no longer have the source code and need to
maintain or extend the compiled code. You can even generate debug symbols for the
compiled code.

Let's take a deeper look at the tools mentioned, starting with code metrics.

[ 359 ]
Using Tools to Improve Code Quality Chapter 12

Performing code cleanup and calculating

code metrics
Before we look at how to gather code metrics, we first need to know what they are and why
they are useful to us. Code metrics are mainly concerned with software complexity and
maintainability. They help us to see how we can improve the maintainability of our source
code and reduce source code complexity.

The code metrics that Visual Studio 2019 calculates for you consist of the following:

Maintainability index: Code maintainability is an essential component

of Application Lifecycle Management (ALM). Until software reaches its end of
life, it must be maintained. The harder the code base is to maintain, the shorter
the lifespan of the source code before a complete replacement is required.
Writing new software to replace an ailing system is far more work and is more
expensive when compared to maintaining an existing system. The measurement
for code maintainability is known as the maintainability index. This value is
an integer value between 0 and 100. Here are the maintainability index ratings,
their colors, and their meanings:
Any value from 20 and above has a green rating for good
maintainability.
Moderately maintainable code is between 10 and 19, with
a yellow rating.
Anything below 10 has a rating of red, meaning that it is hard to
maintain.
Cyclomatic complexity: Code complexity, also known as cyclomatic complexity,
refers to the various code paths through the software. The more paths there are,
the more complex the software is. And the more complex the software is, the
harder it is to test and maintain. Complex code can lead to more error-prone
software releases and can make it hard to maintain and extend the software.
Hence, it is advisable that code complexity should be kept to a minimum.

[ 360 ]
Using Tools to Improve Code Quality Chapter 12

Depth of Inheritance: The Depth of Inheritance and class coupling metrics are
affected by the popular programming paradigm called Object-Oriented
Programming (OOP). With OOP, classes are able to inherit from other classes. A
class that is inherited from is known as a base class. Classes that inherit from a
base class are known as subclasses. The metric for the number of classes that
inherit from each other is known as the Depth of Inheritance.

The deeper the level of inheritance, the more chance you have of errors in derived
classes if something is changed in one of the base classes. The ideal Depth of
Inheritance is 1.

Class coupling: OOP allows class coupling. Class coupling arises when a class is
directly referenced by a parameter, a local variable, a return type, a method call,
a generic or template instantiation, base classes, interface implementations, fields
defined on extra types, and an attribute decoration.

The class coupling code metric determines the level of coupling between classes.
To make code easier to maintain and extend, class coupling should be kept to an
absolute minimum. In OOP, one way to achieve this is to use interface-based
programming. This way, you avoid directly accessing a class. The benefit of this
method of programming is that you can swap classes in and out, as long as they
implement the same interface. Poor-quality code has high coupling and low
cohesion, but good-quality code has low coupling and high cohesion.

Ideally, software should be highly cohesive with low coupling, because it

makes programs easier to test, maintain, and extend.

Lines of source code: The complete count of the lines of your source code,
including blank lines, is measured by the lines of source code metric.
Lines of executable code: The measure of operations in executable code is
measured by the lines of executable code metric.

[ 361 ]
Using Tools to Improve Code Quality Chapter 12

Now that you have a heads-up on what code metrics are and which measurements are
available in Visual Studio 2019 version 16.4 onward, it's time to see them in action, as
follows:

1. Open any project you like within Visual Studio.

2. Right-click on the project.
3. Select Analyze and Code Cleanup | Run Code Cleanup (Profile 1), as illustrated
in the following screenshot:

4. Now, select Calculate Code Metrics.

5. You should see the Code Metrics Results window appear, as shown in the
following screenshot:

[ 362 ]
Using Tools to Improve Code Quality Chapter 12

As you can see from the screenshot, all our classes, interfaces, and methods are marked
with a green indicator. This means that the selected project is one that is maintainable. If any
of these lines were marked yellow or red, then you would need to address them and
refactor them to make them green. Well, we've covered code metrics, and so, naturally, we
move on to cover code analysis.

Performing code analysis

To help developers identify potential problems with their source code, Microsoft provides
the Code Analysis tool as part of Visual Studio. Code Analysis performs a static source
code analysis. The tool will identify design flaws, issues with globalization, security
problems, issues with performance, and interoperability problems.

[ 363 ]
Using Tools to Improve Code Quality Chapter 12

Open the book solution, and select the CH11_AddressingCrossCuttingConcerns project.

Then, from the Project menu, select Project |
CH11_AddressingCrossCuttingConcerns | Properties from the menu. On the properties
page for the project, select Code Analysis, as illustrated in the following screenshot:

As shown in the preceding screenshot, if you see that the recommended analyzer package
is not installed, click on Install to install it. Once installed, the version number will be
displayed in the installed version box. For me, it is version 2.9.6. By default, the active rules
are Microsoft Managed Recommended Rules. The location of this ruleset, as shown in the
description, is C:\Program Files (x86)\Microsoft Visual Studio\2019\Professional\Team
Tools\Static Analysis Tools\Rule Sets\MinimumRecommendedRules.ruleset. Open the
file. It will open as a Visual Studio tool window, as shown here:

[ 364 ]
Using Tools to Improve Code Quality Chapter 12

As you can see in the preceding screenshot, you can select and deselect rules. When you
close the window, you will be prompted to save any changes. To run a code analysis, go
to Analyze and Code Cleanup | Code Analysis. In order to view the results, you will need
the Error List window to be open. You can open it from the View menu.

Once you have run the code analysis, you will see a list of errors, warnings, and messages.
You can address each and every one of them to improve the overall quality of your
software. A sample of these can be seen in the following screenshot:

[ 365 ]
Using Tools to Improve Code Quality Chapter 12

From the preceding screenshot, you can see that the

CH10_AddressingCrossCuttingConcerns project has 32 warnings and 13 messages. If we
were to work on the warnings and messages, we would get them down to 0 messages and 0
warnings. So, now that you have seen how to use code metrics to see how maintainable
your software is and you've analyzed it to see what improvements you can make, it's now
time to look at quick actions.

Using quick actions

Another handy tool that I like to use is the Quick Action tool. Appearing as a screwdriver
, a lightbulb , or an error light bulb on a line of code, quick actions enable you to
use a single command that will generate code, refactor code, suppress warnings, perform
code fixes, and add using statements.

Since the CH10_AddressingCrossCuttingConcerns project had 32 warnings and 13

messages, we can use this project to see the quick actions in action. Have a look at the
following screenshot:

Looking at the preceding screenshot, we see the lightbulb on line 10. If we click on the
lightbulb, the following menu pops up:

[ 366 ]
Using Tools to Improve Code Quality Chapter 12

If we click on Add readonly modifier, the readonly access modifier is placed after the
private access modifier. Have a go yourself at using quick actions to modify the code. It is
fairly straightforward once you get the hang of it. Once you have had a play around with
quick actions, move on to look at the JetBrains dotTrace profiling tool.

Using the JetBrains dotTrace profiler

The JetBrains dotTrace profiler is a part of JetBrains ReSharper Ultimate license. Since we
will be looking at both tools, I recommend that you download and install JetBrains
ReSharper Ultimate before we continue.

JetBrains does have a trial version available if you don't already own a
copy. There are versions available for Windows, macOS, and Linux.

The JetBrains dotTrace profiling tool works with Mono, .NET Framework, and .NET Core.
All application types are supported by the profiler, and you can use the profiler to analyze
and track down performance issues with your code base. The profiler will help you to get
to the bottom of such problems that cause 100% CPU usage, 100% of the disk I/O, maxing
out the memory or running into overflow exception, and many other issues.

Many applications perform HyperText Transfer Protocol (HTTP) requests. The profiler
will analyze how the application is processing these requests, and it will also do the same
with Structured Query Language (SQL) queries on a database. Static methods and unit
tests can be profiled, and you can view the results from within Visual Studio. There is also a
standalone version that you can use.

There are four basic profiling options—Sampling, Tracing, Line-by-Line, and Timeline.
The first time you start looking at the performance of an application, you may decide to use
Sampling, which provides an accurate measurement of call time. Tracing and Line-by-
Line offer more detailed profiling, but they do add more overhead (memory and CPU
usage) to the program being profiled. Timeline is similar to sampling and collects
application events over time. Between them, there is no problem that can't be tracked down
and resolved.

Advanced profiling options include real-time performance counters, thread time, real-time
CPU instructions, and thread cycle time. The real-time performance counters measure the
time between method entry and exit. Thread time measures the thread running time. Based
on the CPU register, the real-time CPU instructions provide an accurate time of method
entry and exit.

[ 367 ]
Using Tools to Improve Code Quality Chapter 12

The profiler can attach to running .NET Framework 4.0 (or later) or .NET Core 3.0 (or later)
applications and processes, profile local applications, and profile remote applications.
These include standalone applications; .NET Core applications; Internet Information
Services (IIS)-hosted web applications; IIS Express-hosted applications; .NET Windows
Services, Windows Communication Foundation (WCF) services; Windows Store and
Universal Windows Platform (UWP) applications; any .NET processes (started after you
run the profiling session); desktop or console applications based on Mono; and Unity editor
or standalone Unity applications.

To access the profiler in Visual Studio 2019 from the menu, select Extensions | ReSharper |
Profile | Show Performance Profiler. In the following screenshot, you can see that nothing
has been profiled yet. Also, the currently selected project to be profiled is set to Basic CH3,
and the profiling type is set to Timeline. We will profile CH3 using Sampling to profile our
project by expanding the Timeline drop-down functionality and selecting Sampling, as
illustrated in the following screenshot:

If you want to sample a different project, just expand the Project drop-down list and select
the project that you want to profile. The project will be built, and the profiler started. Your
project will then run and be shut down. The results will then be displayed in the dotTrace
profiling application, as shown in the following screenshot:

[ 368 ]
Using Tools to Improve Code Quality Chapter 12

From the preceding screenshot, you can see that the first of four threads are being shown.
This is the thread for our program. The other threads are for the supporting processes that
enable our program to run along with the finalizer thread that is responsible for exiting the
program and cleaning up system resources.

The All Calls menu items down the left-hand side comprise the following:

Thread Tree
Call Tree
Plain List
Hot Spots

The current option selected the Thread Tree. Let's have a look at the expanded Call Tree in
the following screenshot:

[ 369 ]
Using Tools to Improve Code Quality Chapter 12

The profiler shows you the complete Call Tree for your code, and that includes system
code as well as your own code. You can see the percentage of time spent on making the call.
This allows you to identify any long-running methods and address them.

Now, we'll look at the Plain List. As you can see with the Plain List view in the screenshot
that follows, we can group it according to the following criteria:

None
Class
Namespace
Assembly

You can see the preceding criteria in the following screenshot:

[ 370 ]
Using Tools to Improve Code Quality Chapter 12

When you click on an item in the list, you get to view the source code for the class where
the method resides. This is useful, as you can see the code where the problem lies and what
needs to be done. The last sampling profile screen we'll look at is the Hot Spots view,
illustrated in the following screenshot:

The profiler is showing that the Main Thread, which is our code's starting point, only takes
4.59% of the processing time. If you click on the root, 18% of the code is our user code, and
72% of the code is system code, as shown in the following screenshot:

[ 371 ]
Using Tools to Improve Code Quality Chapter 12

We have only touched the surface with this profiling tool. There is more to it, and I
encourage you to try it out for yourself. The main purpose of this chapter is to introduce
you to the tools that are available to you.

For further information on how to use JetBrains dotTrace, I refer you to

their online learning materials, at https://www.jetbrains.com/profiler/
documentation/documentation.html.

Next up, we look at JetBrains ReSharper.

Using JetBrains ReSharper

In this section, we look at how JetBrains ReSharper can help you improve your code.
ReSharper is quite an extensive tool, and just as with the profiler, which is a part of the
Ultimate edition of ReSharper, we will only be touching the surface, but you will hopefully
come to an appreciation of what the tool is and what it can do for you to improve your
Visual Studio coding experience. Here are a few benefits of using ReSharper:

With ReSharper, you can perform an analysis of your code quality.

It will provide options to improve your code, remove code smells, and fix coding
problems.
With the navigation system, you are able to completely traverse your solution
and jump to any item of interest. You have many different helpers that include
extended IntelliSense, code reorganization, and more.
Refactoring benefits from ReSharper's offerings that can be localized or solution-
wide.
You can also generate source code using ReSharper, such as base class and
superclasses, and inline methods.
Here, code can be cleaned up in keeping with your company's coding policies to
get rid of unused imports and other unused code.

[ 372 ]
Using Tools to Improve Code Quality Chapter 12

You can visit the ReSharper menu from the Visual Studio 2019 Extensions menu. When in
the code editor, right-clicking the mouse on a piece of code will bring up a context menu
with the appropriate menu items. The ReSharper menu item in the context menu is
Refactor This..., as shown in the following screenshot:

[ 373 ]
Using Tools to Improve Code Quality Chapter 12

Now, from the Visual Studio 2019 menu, run Extensions | ReSharper | Inspect | Code
Issues in Solution. ReSharper will process the solution and then display the Inspection
Results window, as shown in the following screenshot:

As you can see in the preceding screenshot, ReSharper found 527 issues with our code—436
of which are being displayed. These problems include common practices and code
improvements, compiler warnings, constraint violations, language usage opportunities,
potential code quality issues, redundancies in code, redundancies in symbol declarations,
spelling issues, and syntax style.

If we expand Compiler Warnings, we see that there are three problems, as follows:

The _name field is never assigned.

The nre local variable is never used.
This async method lacks await operators and will run synchronously. Use the
await operator to await non-blocking Application Programming Interface (API)
calls, or await TaskEx.Run(...) to do CPU-bound work on a background
thread.

These problems are variable declarations that don't get assigned or used, and an async
method lacking an await operator that will run synchronously. If you click on the first
warning, it will take you to the line of code that is never assigned. Looking at the class, you
can see that the string is declared and used, but it is never assigned. Since we check if the
string contains string.Empty, we can assign that value to the declaration. Hence, the
updated line will be as follows:
private string _name = string.Empty;

[ 374 ]
Using Tools to Improve Code Quality Chapter 12

Since the _name variable still highlights, we can hover over it and see what the problem is.
The Quick Action informs us that the _name variable can be marked read-only. Let's add
the readonly modifier. So, the line now becomes this:
private readonly string _name = string.Empty;

If we click on the refresh button, we will find that the number of issues found is now
526. Yet, we fixed two problems. So, should the number be 525? Well, no. The second
problem that we fixed was not a problem picked up by ReSharper, but an improvement
picked up by Visual Studio Quick Actions. So, ReSharper is showing the correct number of
issues it has detected.

Let's have a look at the potential code quality issue for the LooseCouplingB class.
ReSharper reports a possible System.NullReferenceException within this method.
Let's look at the code first, as follows:
public LooseCouplingB()
{
LooseCouplingA lca = new LooseCouplingA();
lca = null;
Debug.WriteLine($"Name is {lca.Name}");
}

And sure enough, we do have System.NullReferenceException staring us in the face.

We'll look at the LooseCouplingA class, to confirm which members should be set to null.
Also, the member to be set is _name, as illustrated in the following code snippet:
public string Name
{
get => _name.Equals(string.Empty) ? StringIsEmpty : _name;

set
{
if (value.Equals(string.Empty))
Debug.WriteLine("Exception: String length must be greater than
zero.");
}
}

[ 375 ]
Using Tools to Improve Code Quality Chapter 12

However, _name is being checked for empty. And so, really, the code should be setting
_name to string.Empty. So, our fixed constructor in LooseCouplingB becomes the
following:
public LooseCouplingB()
{
var lca = new LooseCouplingA
{
Name = string.Empty
};
Debug.WriteLine($"Name is {lca.Name}");
}

Now, if we refresh the Inspection Results window, our list of issues has gone down by
five, because apart from correctly assigning the Name property, we made use of the
language usage opportunity to simplify our instantiation and initialization, which was
detected by ReSharper. Have a play around with the tool and eliminate the problems found
in the Inspection Results window.

ReSharper can also generate dependency diagrams. To generate a dependency diagram for
our solution, select Extensions | ReSharper | Architecture | Show Project Dependency
Diagram. This will display the project dependency diagram for our solution. The black
container box called CH06 is the namespace, and the gray/blue boxes prefixed with CH06_
are projects, as illustrated in the following screenshot:

[ 376 ]
Using Tools to Improve Code Quality Chapter 12

As you can see from the project dependency diagram in the CH06 namespace, there is a
project dependency between CH06_SpecFlow and CH06_SpecFlow.Implementation.
Similarly, you can also generate type dependency diagrams using ReSharper. Select
Extensions | ReSharper | Architecture | Type Dependencies Diagram.

If we generate the diagram for ConcreteClass in the

CH10_AddressingCrossCuttingConcerns project, then the diagram will be generated,
but only the ConcreteComponent class will be initially displayed. Right-click the
ConcreteComponent box on the diagram and select Add All Referenced Types. You will
see the addition of the ExceptionAttribute class and the IComponent interface. Right-
click on the ExceptionAttribute class and select Add All Referenced Types, and you
end up with the following:

[ 377 ]
Using Tools to Improve Code Quality Chapter 12

What's really wonderful about this tool is that you can order the diagram elements by
namespace. This can be really useful for massive solutions with multiple large projects and
deep-nested namespaces. Though it's good that we can right-click on code and go to the
item declaration, you can't beat visually seeing the lay of the land in terms of the project
that you are working on, and that is why this tool can be really useful. Here is an example
of a typed dependencies diagram organized by namespaces:

Many a time, I could have really used a diagram such as this in my day-to-day work. This
diagram is technical documentation that will help developers find their way around a
complex solution. They will be able to see which namespaces are available and how
everything is interlinked. This will empower developers with the correct knowledge as to
where new classes, enums, and interfaces should be placed when performing new
development, but also, they will know where to find objects if they are performing
maintenance. This diagram is also good for finding duplicate namespaces, interfaces, and
object names.

[ 378 ]
Using Tools to Improve Code Quality Chapter 12

Let's now look at coverage. Proceed as follows:

1. Select Extensions | ReSharper | Cover | Cover Application.

2. The Coverage Configuration dialog will be displayed, and the default selected
option will be Standalone.
3. Select your executable.
4. You can select a .NET app from the bin folder.
5. The following screenshot shows the Coverage Configuration dialog:

[ 379 ]
Using Tools to Improve Code Quality Chapter 12

6. Click the Run button to start the application and collect profiling data. ReSharper
will display the following dialog:

The application will then run. As the application is running, the coverage profiler
will be collecting data. Our selected executable is a console application that
displays the following data:

7. Click the console window, and then press any key to exit. The coverage dialog
will disappear, and storage will then be initialized. Finally, the Coverage Results
Browser window will be displayed, as shown here:

[ 380 ]
Using Tools to Improve Code Quality Chapter 12

This window contains really useful information. It provides a visual indicator of code that
was not called, marked in red. The code that was executed is marked in green. Using this
information, you can see if the code is dead code that can be removed, or was not executed
due to the path taken through the system but is still required, or was commented out for
testing purposes, or was simply not called because the developer forgot to add the call in
the correct place or a condition check was wrong.

[ 381 ]
Using Tools to Improve Code Quality Chapter 12

To go to the item of interest, you just have to double-click on the item, and then you will be
taken to the specific code you are interested in. Our Program class only covers 33% of the
code. So, let's double-click Program, and see what's the matter. The resulting output is
shown in the following code block:
static void Main(string[] args)
{
LoggingServices.DefaultBackend = new ConsoleLoggingBackend();
AuditServices.RecordPublished += AuditServices_RecordPublished;
DecoratorPatternExample();
//ProxyPatternExample();
//SecurityExample();

//ExceptionHandlingAttributeExample();

//SuccessfulMethod();
//FailedMethod();

Console.ReadKey();
}

As you can see from the code, the reason why some of our code was not covered is because
calls to the code were commented out for testing purposes. We can leave the code as it is
(which we will do in this case). However, you can also remove the dead code or reinstate
the code by removing the comments. Now, you know why the code is not being covered.

Well, now that you've been introduced to ReSharper and had a look at tools to assist you in
writing good, clean C# code, it is time to look at our next tool, called Telerik JustDecompile.

Using Telerik JustDecompile

I have used Telerik JustDecompile on a number of occasions, for things such as tracking
down bugs in third-party libraries, recovering essential project source code that has been
lost, checking the strength of assembly obfuscation, and for learning purposes. It is a tool
that I highly recommend, as over the years it has proven its worth many times.

The decompilation engine is open source and you can obtain the source code from https:/
/github.com/telerik/justdecompileengine, so you are free to contribute to the project
and write your own extensions for it. You can download Windows Installer from the
Telerik website, at https://www.telerik.com/products/decompiler.aspx. All source code
is fully navigable. The decompiler is available as a standalone application or as a Visual
Studio extension. You create VB.NET or C# projects from assemblies that you decompile,
and you extract and save resources from the decompiled assemblies.

[ 382 ]
Using Tools to Improve Code Quality Chapter 12

Download and install Telerik JustDecompile. We will then go through the decompilation
process, and generate a C# project from an assembly. You may be prompted to install other
tools during the installation process, but you can deselect the other offerings from Telerik.

Run the Telerik JustDecompile standalone application. Find a .NET assembly and then drag
it into the left pane of Telerik JustDecompile. It will decompile the code and display the
code tree on the left. If you select an item on the left, the code is shown on the right, as you
can see in the screenshot:

As you can see, the decompilation process is fast and it does a pretty good job of
decompiling our assembly. The decompilation is not perfect, but in most cases, it does the
job. Proceed as follows:

1. In the dropdown to the right of the Plugins menu item, select C#.
2. Then, click on Tools | Create Project.
3. You will sometimes be prompted to select the .NET version to target; other times,
not.
4. Then, you will be asked where to save the project.
5. The project will then be written to that location.

[ 383 ]
Using Tools to Improve Code Quality Chapter 12

You can then open the project in Visual Studio and work on it. Should you encounter any
problems, Telerik logs the issues in your code and provides an email. You can always email
them with any issues you encounter. They are good at responding to and fixing problems.

Well, we have completed our look at the tools in this chapter, so now, let's look at what we
have learned in summary.

Summary
In this chapter, you have seen how code metrics provide several measurements of code
quality, and how easy it is to generate them. Code metrics include the number of
lines—including blank lines—versus the number of executable lines of code, the cyclomatic
complexity, the level of cohesion and coupling, and how maintainable your code is. The
refactoring color codes are green for good, yellow for ideally needs refactoring, and red for
definitely needs refactoring.

You then saw how easy it is to provide a static code analysis of projects and view the
results. Viewing and modifying rulesets that govern what gets analyzed and what doesn't
get analyzed was also covered. Then, you experienced quick actions and saw how we can
perform bug fixes, add using statements, and refactor code with a single command.

We then used the JetBrains dotTrace profiler to measure our application's performance,
track down bottlenecks, and identify hungry methods that take up the most processing
time. The next tool we looked at was JetBrains ReSharper, which enables us to inspect code
for various problems and potential improvements. We identified a couple of them and
made the necessary changes, and saw how easy it was to improve the code with this tool.
Then, we looked at creating architectural diagrams for dependencies and type
dependencies.

Finally, we looked at Telerik JustDecompile, a very useful tool that can be used to
decompile assemblies and generate projects in either C# or VB.NET from them. This can be
very useful when bugs are encountered or the program needs to be expanded, but you no
longer have access to the existing source code.

In the chapters that follow, we will mainly be looking at code, and how we can refactor it.
But for now, test your knowledge with the following questions and further your reading
with the links provided in the Further reading section.

[ 384 ]
Using Tools to Improve Code Quality Chapter 12

Questions
1. What are code metrics, and why should we use them?
2. Name six code metric measurements.
3. What is code analysis, and why is it useful?
4. What are quick actions?
5. What is JetBrains dotTrace used for?
6. What is JetBrains ReSharper used for?
7. Why use Telerik JustDecompile to decompile assemblies?

Further reading
Official Microsoft documentation on code metrics: https://docs.microsoft.
com/en-us/visualstudio/code-quality/code-metrics-values?view=vs-2019
Official Microsoft documentation on Quick Actions: https://docs.microsoft.
com/en-us/visualstudio/ide/quick-actions?view=vs-2019
JetBrains dotTrace profiler: https://www.jetbrains.com/profiler/

[ 385 ]
13
Refactoring C# Code –
Identifying Code Smells
In this chapter, we will look at problem code and how to refactor it. In the industry,
problem code is normally termed code smell. It is code that compiles, runs, and does what
it is supposed to do. The reason it is problem code is that it becomes unreadable, complex
in nature, and makes the code base hard to maintain and extend further down the line.
Such code should be refactored as soon as it's feasible to do so. It is technical debt, and in
the long run, if you don't deal with it, it will bring the project to its knees. When this
happens, you are looking at an expensive redesign and recoding of the application from
scratch.

So what is refactoring? Refactoring is the process of taking existing code that works and
rewriting it such that the code becomes clean. And as you have already discovered, clean
code is easy to read, easy to maintain, and easy to extend.

In this chapter, we will cover the following topics:

Identifying application-level code smells and how we can address them

Identifying class-level code smells and how we can address them
Identifying method-level code smells and how we can address them

After working your way through this chapter, you will have gained the following skills:

Identifying different kinds of code smell

Understanding why the code is classed as code smell
Refactoring code smells so they become clean code

We'll start our look at refactoring code smells by looking at application-level code smells.
Refactoring C# Code – Identifying Code Smells Chapter 13

Technical requirements
You will need the following prerequisites for the chapter:

Visual Studio 2019

PostSharp

For the code files of the chapter, you can use this link: https://github.com/
PacktPublishing/Clean-Code-in-C-/tree/master/CH13.

Application-level code smells

Application-level code smells are problem code scattered through the application and affect
every layer. No matter what layer of the software you find yourself in, you will see the
same problematic code appearing over and over again. If you don't address these issues
now, then you will find that your software will start to die a slow and agonizing death.

In this section, we will look at the application-level code smells and how we can remove
them. Let's start with Boolean blindness.

Boolean blindness
Boolean data blindness refers to the information loss as determined by functions that work
on Boolean values. Using a better structure provides better interfaces and classes that keep
data, making for a more pleasant experience in working with data.

Let's look at the problem of Boolean blindness via this code sample:
public void BookConcert(string concert, bool standing)
{
if (standing)
{
// Issue standing ticket.
}
else
{
// Issue sitting ticket.
}
}

[ 387 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

This method takes a string for the concert name, and a Boolean value indicating whether
the person is standing or seated. Now, we would call the code as follows:
private void BooleanBlindnessConcertBooking()
{
var booking = new ProblemCode.ConcertBooking();
booking.BookConcert("Solitary Experiments", true);
}

If someone new to the code saw the BooleanBlindnessConcertBooking() method, do

you think they would know instinctively what true stands for? I think not. They would be
blind to what it means. So they would have to either use IntelliSense or locate the method
being referred to find the meaning. They are Boolean blind. So how can we cure them of
this blindness?

Well, a simple solution would be to replace the Boolean with an enum. Let's start by adding
our enum called TicketType:
[Flags]
internal enum TicketType
{
Seated,
Standing
}

Our enum identifies two types of ticket types. These are Seated and Standing. Now let's
add our ConcertBooking() method:
internal void BookConcert(string concert, TicketType ticketType)
{
if (ticketType == TicketType.Seated)
{
// Issue seated ticket.
}
else
{
// Issue standing ticket.
}
}

The following code shows how to call the newly refactored code:
private void ClearSightedConcertBooking()
{
var booking = new RefactoredCode.ConcertBooking();
booking.BookConcert("Chrom", TicketType.Seated);
}

[ 388 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

Now, if that new person came along and looked at this code, they would see that we are
booking a concert to see the band Chrom, and that we want seated tickets.

Combinatorial explosion
Combinatorial explosion is a by-product of the same thing being performed by different
pieces of code using different combinations of parameters. Let's look at an example that
adds numbers:
public int Add(int x, int y)
{
return x + y;
}

public double Add(double x, double y)

{
return x + y;
}

public float Add(float x, float y)

{
return x + y;
}

Here, we have three methods that all add numbers. The return types and parameters are all
different. Is there a better way? Yes, through the use of generics. By using generics, you can
have one single method that is capable of working with different types. And so, we will be
using generics to solve our addition problem. This will allow us to have a single addition
method that will accept either integers, doubles, or floats. Let's have a look at our new
method:
public T Add<T>(T x, T y)
{
dynamic a = x;
dynamic b = y;
return a + b;
}

[ 389 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

This generic method is called with a specific type assigned to T. It performs the addition
and returns the result. Only one version of the method is required for the different .NET
types that can be added together. To call the code for int, double, and float values, we
would do the following:
var addition = new RefactoredCode.Maths();
addition.Add<int>(1, 2);
addition.Add<double>(1.2, 3.4);
addition.Add<float>(5.6f, 7.8f);

We have just eliminated three methods and replaced them with a single method that
performs the same task.

Contrived complexity
When you can develop code with simple architecture, but instead implement an advanced
and rather complex architecture, this is known as contrived complexity. Unfortunately, I
have suffered having to work on such systems and it is a proper pain and cause of stress.
What you find with such systems is that they tend to have a high turnover of staff. They
lack documentation, and no one seems to know the system or has the ability to answer
questions by onboarders—the poor souls who have to learn the system to maintain and
extend it.

My advice to all super-intelligent software architects is that when it comes to software,

Keep It Simple, Stupid (KISS). Remember, the days of permanent employment with jobs
for life appear to be a thing of the past now. Oftentimes, programmers are more for chasing
the money than showing lifelong loyalty to the business. So with the business relying on
the software for revenue, you need a system that is easy to understand, to onboard new
staff, to maintain, and to extend. Ask yourself this question: If the systems that you are
responsible for suddenly experienced yourself and all staff assigned to them walking out
and finding new opportunities, would the new staff who take over be able to hit the ground
running? Or would they be left stressed out and scratching their heads?

Also bear in mind that if you have only one person on the team who understands that
system and they die, move on to a new location, or retire, where does that leave you and
the rest of the team? And even more than that, where does it leave the business?

I cannot stress enough that you really are to KISS. The only reason for creating complex
systems and not documenting them and sharing the architectural knowledge is to hold the
business over a barrel so they keep you on and you can bleed them dry. Don't do it. In my
experience, the more complicated a system is, the quicker it dies a death and has to be
rewritten.

[ 390 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

In Chapter 12, Using Tools to Improve Code Quality, you learned how to use Visual Studio
2019 tools to discover the cyclomatic complexity and Depth of Inheritance. You also learned
how to produce dependency diagrams with ReSharper. Use these tools to discover problem
areas in the code, then focus on those areas. Reduce cyclomatic complexity down to a value
of 10 or less. And reduce the depth of inheritance on all objects down to no greater than 1.

Then, make sure all classes only perform the tasks that they are meant to. Aim to keep
methods small. A good rule of thumb is to have no more than around 10 lines of code per
method. As for method parameters, replace long parameter lists with parameter objects.
And where you have a lot of out parameters, refactor the method to return a tuple or
object. Identify any multithreading, and make sure that the code being accessed is thread-
safe. You have seen in Chapter 9, Designing and Developing APIs, how to replace mutable
objects with immutable ones to improve thread-safety.

Also, look for the Quick Tips icons. These will normally suggest one-click refactorings for
the line of code they highlight. I recommend you use them. These were mentioned in
Chapter 12, Using Tools to Improve Code Quality.

The next code smell to consider is the data clump.

Data clump
A data clump occurs when you see the same fields appearing together in different classes
and parameter lists. Their names usually follow the same pattern. This is normally the sign
that a class is missing from the system. The reduction in system complexity will come by
identifying the missing class and generalizing it. Don't be put off by the fact that the class
may only be small, and never think of a class as being unimportant. If there is a need for a
class to simplify the code, then add it.

Deodorant comments
When a comment uses nice words to excuse bad code, this is known as a deodorant
comment. If the code is bad, then refactor it to make it good and remove the comment. If
you don't know how to refactor it to make it good, then ask for help. If there is no one to
ask that can help you, then post your code on Stack Overflow. There are some very good
programmers on that site that can be a real help to you. Just make sure to follow the rules
when posting!

[ 391 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

Duplicate code
Duplicate code is code that occurs more than once. Problems that arise from duplicate code
include increased maintenance cost per duplication. When a developer is fixing a piece of
code, it costs the business time and money. Fixing 1 bug is technical debt (programmer's pay) x
1. But if there are 10 duplications of that code, that's technical debt x 10. So the more that
code is duplicated, the more expensive it is to maintain. Then there is the boredom factor of
having to fix the same problem in multiple locations. And the fact that duplication may get
overlooked by the programmer doing the bug fix.

It is best to refactor the duplicate code so that only one copy of the code exists. Often, the
easiest way to do this is to add it to a new reusable class in your current project and place it
in a class library. The benefit of placing reusable code in a class library is that other projects
can use the same file.

In the present day, it is best to use the .NET Standard class library for
building reusable code. The reason for this is that .NET Standard libraries
can be accessed by all C# project types on Windows, Linux, macOS, iOS,
and Android.

Another alternative for removing boilerplate code is to use Aspect-Oriented Programming

(AOP). We looked at AOP in the previous chapter. You essentially move boilerplate code
into an aspect. The aspect then decorates the method it is applied to. When the method is
compiled, the boilerplate code is then weaved into place. This enables you only to write
code that meets the business requirement inside the method. The aspect applied to the
method hides the code that is essential, but not part of what the business has asked for. This
coding technique is nice and clean, and it works really well.

You can also write decorators using the decorator pattern, as you also saw in the previous
chapter. The decorator wraps concrete class operations in such a way that you can add new
code without affecting the expected operation of the code. A simple example would be to
wrap the operation in a try/catch block as you saw previously in Chapter 11, Addressing
Cross-Cutting Concerns.

Lost intent
If you can't easily understand the intent of the source code, then it has lost its intent.

The first thing to do is look at the namespace and the class name. These should indicate the
purpose of the class. Then, check the contents of the class, and look for code that looks out
of place. Once you have identified such code, refactor the code and place it in the right
location.

[ 392 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

The next thing to do is to look at each of the methods. Are they only doing one thing well or
doing multiple things not so well? If yes, then refactor them. For large methods, look for
code that can be extracted out into a method. Aim to make the code of the class read like a
book. Keep refactoring the code until the intent is clear, and only what is in the class needs
to be in the class.

Don't forget to put the tools to work that you learned how to use in Chapter 12, Using Tools
to Improve Code Quality. The mutation of variables is the code smell we will look at next.

The mutation of variables

The mutation of variables means they are hard to understand and reason about. This makes
them difficult to refactor.

A mutable variable is one that gets changed multiple times by different operations. This
makes reasoning about why is the value more difficult. Not only that, but because the
variable is mutating from different operations, this makes it difficult to extract sections of
code into other small and more readable methods. Mutable variables can also require more
checking that adds complexity to the code.

Look to refactor small sections of code by extracting them out to methods. If there is a lot of
branching and looping, see if there is an easier way to do things to remove the complexity.
If you are using multiple out values, consider returning an object or tuple. Aim to remove
the mutability of the variable to make it easier to reason about, and know why it is the
value that it is, and from where it is getting set. Remember that the smaller the method is
that holds a variable, the easier it will be to determine where the variable is getting set, and
why.

Look at the following example:

[InstrumentationAspect]
public class Mutant
{
public int IntegerSquaredSum(List<int> integers)
{
var squaredSum = 0;
foreach (var integer in integers)
{
squaredSum += integer * integer;
}
return squaredSum;
}
}

[ 393 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

The method takes a list of integers. It then loops through the integers, squares them, and
then adds them to the squaredSum variable that is returned when the method exits. Notice
the iterations, and the fact that the local variable is getting updated in each iteration. We
can improve on this using LINQ. The following code shows the improved, refactored
version:
[InstrumentationAspect]
public class Function
{
public int IntegerSquaredSum(List<int> integers)
{
return integers.Sum(integer => integer * integer);
}
}

In our new version, we use LINQ. As you know from an earlier chapter, LINQ employs
functional programming. As you can see here, there is no loop, and no local variable being
mutated.

Compile and run the program, and you will see the following:

Both versions of the code produce the same output.

You will have noticed that both versions of the code have [InstrumentationAspect]
applied to them. We added this aspect to our reusable library in Chapter 12, Addressing
Cross-Cutting Concerns. When you run the code, you will find a Logs folder in the Debug
folder. Open the Profile.log file in Notepad, and you will see the following output:
Method: IntegerSquaredSum, Start Time: 01/07/2020 11:41:43
Method: IntegerSquaredSum, Stop Time: 01/07/2020 11:41:43, Duration:
00:00:00.0005489
Method: IntegerSquaredSum, Start Time: 01/07/2020 11:41:43
Method: IntegerSquaredSum, Stop Time: 01/07/2020 11:41:43, Duration:
00:00:00.0000027

The output shows that the ProblemCode.IntegerSquaredSum() method was the slowest
version, taking 548.9 nanoseconds to run. And that the
RefactoredCode.IntegerSquaredSum() method was much faster, taking only 2.7
nanoseconds to run.

[ 394 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

By refactoring the loop to use LINQ, we avoided mutating a local variable. And we also
reduced the time it took to process the calculation by 546.2 nanoseconds. Such a small
improvement is not noticeable to the human eye. But if you perform such calculations on
big data, then you will experience a noticeable difference.

We'll now discuss the oddball solution.

The oddball solution

When you see a problem solved in a different way throughout the source code, this is
known as an oddball solution. This can happen because of different programmers having
their own style of programming, and no standards being put in place. It can also happen
through ignorance of the system, in that the programmer does not realize a solution already
exists.

A way to refactor oddball solutions is to write a new class that encompasses the behavior
that is being repeated in different ways. Add the behavior to the class in the cleanest way
that is the most performant. Then, replace the oddball solutions with the newly refactored
behavior.

You can also unite different system interfaces using the Adapter Pattern:

[ 395 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

The Target class is the domain-specific interface that is used by Client. An existing
interface that needs adapting is called Adaptee. The Adapter class adapts the Adaptee
class to the Target class. And finally, the Client class communicates objects that conform
to the Target interface. Let's implement the adapter pattern. Add a new class called
Adaptee:

public class Adaptee

{
public void AdapteeOperation()
{
Console.WriteLine($"AdapteeOperation() has just executed.");
}
}

The Adaptee class is very simple. It contains a method called AdapteeOperation() that
prints out a message to the console. Now add the Target class:
public class Target
{
public virtual void Operation()
{
Console.WriteLine("Target.Operation() has executed.");
}
}

The Target class is also very simple and contains a virtual method called Operation()
that prints out a message to the console. We'll now add the Adapter class that wires
Target and Adaptee together:

public class Adapter : Target

{
private readonly Adaptee _adaptee = new Adaptee();

public override void Operation()

{
_adaptee.AdapteeOperation();
}
}

The Adapter class inherits the Target class. We then create a member variable to hold our
Adaptee object and initialize it. We then have a single method that is the overridden
Operation() method of the Target class. Finally, we will add our Client class:

public class Client

{
public void Operation()

[ 396 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

{
Target target = new Adapter();
target.Operation();
}
}

The Client class has a single method called Operation(). This method creates a new
Adapter object and assigns it to a Target variable. It then calls the Operation() method
on the Target variable. If you call a new Client().Operation() method and run the
code, you will see the following output:

You can see from the screenshot that the method that gets executed is the
Adaptee.AdapteeOperation() method. Now that you have successfully learned how to
implement the adapter pattern to solve oddball solutions, we will move on to look at
shotgun surgery.

Shotgun surgery
Making a single change that requires changes to multiple classes is known as shotgun
surgery. This can sometimes be down to excessive refactoring of code due to divergent
changes being encountered. This code smell increases the propensity for introducing bugs
such as those caused by a missed chance. You also increase the possibility of merge
conflicts, because the code needs to change in so many areas that programmers end up
stepping on each other's toes. The code is that convoluted that it induces cognitive overload
in programmers. And new programmers have a steep learning curve because of the nature
of the software.

The version control history will provide a history of the changes made to the software over
time. This can help you identify all the areas that are changed, every time a new piece of
functionality is added or when a bug is encountered. Once these areas have been identified,
then you can look to move the changes to a more localized area of the code base. This way,
when a change is required, you only have to focus on one area of the program and not
many areas. This makes the maintenance of the project a lot easier.

[ 397 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

Duplicate code is a good candidate for refactoring into a single class that is appropriately
named, and that is placed in the correct namespace. Also, consider all the different layers of
your application. Are they really necessary? Can things be simplified? In a database-driven
application, is it really necessary to have DTOs, DAOs, domain objects, and the like? Could
database access be simplified in any way? These are just some ideas for reducing the size of
the code base, and so reducing the number of areas that must be modified to effect a
change.

Other things to look at are the level of coupling and cohesion. Coupling needs to be kept to
an absolute minimum. One way to accomplish this is to inject dependencies via
constructors, properties, and methods. The injected dependencies would be of a specific
interface type. We will code a simple example. Add an interface called IService:
public interface IService
{
void Operation();
}

The interface contains a single method called Operation(). Now, add a class called
Dependency that implements IService:

public class Dependency : IService

{
public void Operation()
{
Console.WriteLine("Dependency.Operation() has executed.");
}
}

The Dependency class implements the IService interface. In the Operation() method, a
message is printed to the console. Now let's add the LooselyCoupled class:
public class LooselyCoupled
{
private readonly IService _service;

public LooselyCoupled(IService service)

{
_service = service;
}

public void DoWork()

{
_service.Operation();
}
}

[ 398 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

As you can see, the constructor takes a type of IService and stores it in a member
variable. The call to DoWork() calls the Operation() method within the IService type.
The LooselyCoupled class is just that loosely coupled, and it is easy to test.

By reducing coupling, you make classes easier to test. By removing code that does not
belong in a class and placing it where it does belong, you improve the readability,
maintainability, and extensibility of the application. You lessen the learning curve for
anyone coming on board, and there is less chance of introducing bugs when you perform
maintenance or new development.

Let's now have a look at solution sprawl.

Solution sprawl
The single responsibility that is implemented within different methods, classes, and even
libraries suffer from solution sprawl. This can make code really hard to read and
understand. The result is that code becomes harder to maintain and extend.

To fix the problem, move the implementation of the single responsibility into the same
class. This way the code is in just one location and does what it needs to. This makes code
easy to read and understand. The result is that the code can be easily maintained and
extended.

Uncontrolled side effects

Uncontrolled side effects are those issues that raise their ugly heads in production because
the quality assurance tests are unable to capture them. When you encounter these
problems, the only option you have is to refactor the code so that it is fully testable and
variables can be viewed during debugging to make sure they are set appropriately.

An example is passing values by reference. Imagine two threads passing a person object by
reference to a method that modifies the person object. A side effect is that unless proper
locking mechanisms are in place, each thread can modify the other thread's person object
invalidating the data. You saw an example of mutable objects in Chapter 8, Threading and
Concurrency.

That concludes our look at application-level code smells. So, now we will move on to look
at class-level code smells.

[ 399 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

Class-level code smells

Class-level code smells are localized problems with the class in question. The kinds of
problems that can plague a class are things like cyclomatic complexity and depth of
inheritance, high coupling, and low cohesion. Your aim when writing a class is to keep it
small and functional. The methods in the class should actually be there, and they should be
small. Only do in the class what needs to be done – no more, no less. Work to remove class
dependency and make your classes testable. Remove code that should be placed elsewhere
to where it belongs. In this section, we address class-level code smells and how to refactor
them, starting with cyclomatic complexity.

Cyclomatic complexity
When a class has a large number of branches and loops, it has an increased cyclomatic
complexity. Ideally, the code should have a cyclomatic complexity value of between 1 and 10.
Such code is simple and without risks. Code with a cyclomatic complexity of 11-20 is
complex but low risk. When the cyclomatic complexity of the code is between 21-50, then
the code requires attention as it is too complex and poses a medium risk to your project.
And if the code has a cyclomatic complexity of more than 50, then such code is high risk
and is not testable. A code that has a value above 50 must be refactored immediately.

The goal of refactoring will be to get the cyclomatic value down to between 1-10. Start by
replacing switch statements followed by if expressions.

Replacing switch statements with the factory pattern

In this section, you will see how to replace a switch statement with the factory pattern.
First, we will need a report enum:
[Flags]
public enum Report
{
StaffShiftPattern,
EndofMonthSalaryRun,
HrStarters,
HrLeavers,
EndofMonthSalesFigures,
YearToDateSalesFigures
}

[ 400 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

The [Flags] attribute enables us to extract the name of the enum. The Report enum
provides a list of reports. Now let's add our switch statement:
public void RunReport(Report report)
{
switch (report)
{
case Report.EndofMonthSalaryRun:
Console.WriteLine("Running End of Month Salary Run Report.");
break;
case Report.EndofMonthSalesFigures:
Console.WriteLine("Running End of Month Sales Figures
Report.");
break;
case Report.HrLeavers:
Console.WriteLine("Running HR Leavers Report.");
break;
case Report.HrStarters:
Console.WriteLine("Running HR Starters Report.");
break;
case Report.StaffShiftPattern:
Console.WriteLine("Running Staff Shift Pattern Report.");
break;
case Report.YearToDateSalesFigures:
Console.WriteLine("Running Year to Date Sales Figures
Report.");
break;
default:
Console.WriteLine("Report unrecognized.");
break;
}
}

Our method accepts a report and then decides on what report to execute. When I started off
in 1999 as a junior VB6 programmer, I was responsible for building a report generator from
scratch for the likes of Thomas Cook, ANZ, BNZ, Vodafone, and a few other big concerns.
There were many reports, and I was responsible for writing a case statement that was
massive that dwarfed this one. But my system worked really well. However, by today's
standards, there are much better ways of performing this same code and I would do things
very differently.

[ 401 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

Let's use the factory method to run our reports without using a switch statement. Add a
file called IReportFactory as shown:
public interface IReportFactory
{
void Run();
}

The IReportFactory interface only has one method called Run(). This method will be
used by the implementing classes to run their reports. We'll only add one report class,
called StaffShiftPatternReport, which implements IReportFactory:
public class StaffShiftPatternReport : IReportFactory
{
public void Run()
{
Console.WriteLine("Running Staff Shift Pattern Report.");
}
}

The StaffShiftPatternReport class implements the IReportFactory interface. The

implemented Run() method prints a message to the screen. Add a report called
ReportRunner:

public class ReportRunner

{
public void RunReport(Report report)
{
var reportName =
$"CH13_CodeRefactoring.RefactoredCode.{report}Report,
CH13_CodeRefactoring";
var factory = Activator.CreateInstance(
Type.GetType(reportName) ?? throw new
InvalidOperationException()
) as IReportFactory;
factory?.Run();
}
}

The ReportRunner class has a method called RunReport. It accepts a parameter of type
Report. With Report being an enum with the [Flags] attribute, we can obtain the name
of the report enum. We use this to build the name of the report. Then, we use the
Activator class to create an instance of the report. If the reportName returns null when
getting the type, InvalidOperationException is thrown. The factory is cast to the
IReportFactory type. We then call the Run method on the factory to generate the report.

[ 402 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

This code is definitely much better than a very long switch statement. We need to know
how to improve the readability of conditional checks within an if statement. We'll look at
that next.

Improving the readability of conditional checks within

an if statement
The if statements can break the single responsibility and the open/closed principles. See
the following example:
public string GetHrReport(string reportName)
{
if (reportName.Equals("Staff Joiners Report"))
return "Staff Joiners Report";
else if (reportName.Equals("Staff Leavers Report"))
return "Staff Leavers Report";
else if (reportName.Equals("Balance Sheet Report"))
return "Balance Sheet Report";
}

The GetReport() class has three responsibilities: the staff joiners report, the staff leavers
report, and the balance sheet report. This breaks the SRP because the method should only
be concerned with HR reports and it is returning HR and Finance reports. As far as the
open/closed principle is concerned, every time a new report is needed we will have to
extend this method. Let's refactor the method so we no longer need the if statement. Add a
new class called ReportBase:
public abstract class ReportBase
{
public abstract void Print();
}

The ReportBase class is an abstract class with an abstract Print() method. We will add
the NewStartersReport class, which inherits the ReportBase class:
internal class NewStartersReport : ReportBase
{
public override void Print()
{
Console.WriteLine("Printing New Starters Report.");
}
}

[ 403 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

The NewStartersReport class inherits the ReportBase class and overrides the Print()
method. The Print() method prints a message to the screen. Now, we will add the
LeaversReport class, which is pretty much the same:

public class LeaversReport : ReportBase

{
public override void Print()
{
Console.WriteLine("Printing Leavers Report.");
}
}

The LeaversReport inherits the ReportBase class and overrides the Print() method.
The Print() method prints a message to the screen. We can now call the reports as
follows:
ReportBase newStarters = new NewStartersReport();
newStarters.Print();

ReportBase leavers = new LeaversReport();

leavers.Print();

Both reports inherit the ReportBase class, and so can be instantiated and assigned to a
ReportBase variable. The Print() method can then be called on the variable, and the
correct Print() method will be executed. The code now adheres to the single
responsibility principle and the open/closed principle.

The next thing we will look at is a divergent change code smell.

Divergent change
When you need to make a change in one location and find yourself having to change many
unrelated methods, then this is known as a divergent change. Divergent changes take place
within a single class and are the result of a poor class structure. Copying and pasting code
is another reason this problem arises.

To fix the problem, move the code causing the problem to its own class. If the behavior and
state are shared between classes, then consider implementing inheritance using base classes
and subclasses as appropriate.

[ 404 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

The benefits of fixing divergent change-related problems include easier maintenance, as

changes will be located within a single location. This makes supporting the application a
whole load easier. It also removes duplicate code from the system, which just so happens to
be the next thing we will be discussing.

Downcasting
When a base class is cast to one of its children, this is known as downcasting. This is clearly
a code smell as the base class should not know about the classes that inherit it. For example,
consider the Animal base class. Any type of animal can inherit the base class. But an animal
can only be of one type. For example, felines are felines and canines are canines. It would be
absurd to cast a feline to a canine and vice versa.

It is even more absurd to downcast an animal to one of its subtypes. That would be like
saying a monkey is the same as a camel and is really good at transporting humans and
cargo long distances through the desert. This just does not make sense. And so, you should
never be downcasting. The upcasting of various animals such as monkeys and camels to the
type Animal is valid because felines, canines, monkeys, and camels are all types of animals.

Excessive literal use

When using literals, it is very easy to introduce coding errors. An example would be a
spelling mistake in a string literal. It is best to assign literals to constant variables. String
literals should be placed in resource files for localization. Especially if you plan to deploy
your software to different locations around the world.

Feature envy
When a method spends more time processing source code in classes other than the one that
it is in, this is known as feature envy. We will see an example of this in our
Authorization class. But before we do, let's have a look at our Authentication class:

public class Authentication

{
private bool _isAuthenticated = false;

public void Login(ICredentials credentials)

{
_isAuthenticated = true;
}

[ 405 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

public void Logout()

{
_isAuthenticated = false;
}

public bool IsAuthenticated()

{
return _isAuthenticated;
}
}

Our Authentication class is responsible for logging people in and out, as well as
identifying whether they are authenticated or not. Add our Authorization class:
public class Authorization
{
private Authentication _authentication;

public Authorization(Authentication authentication)

{
_authentication = authentication;
}

public void Login(ICredentials credentials)

{
_authentication.Login(credentials);
}

public void Logout()

{
_authentication.Logout();
}

public bool IsAuthenticated()

{
return _authentication.IsAuthenticated();
}

public bool IsAuthorized(string role)

{
return IsAuthenticated && role.Contains("Administrator");
}
}

[ 406 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

As you can see with our Authorization class, it is doing more than it is supposed to.
There is one method that validates whether the user is authorized to carry a role. The role
passed in is checked to see whether it is the administrator role. If it is, then the person is
authorized. But if the role is not the administrator role, then the person is not authorized.

However, if you look at the other methods, they are doing no more than calling the same
methods in the Authentication class. So, in the context of this class, the authentication
methods are an example of feature envy. Let's remove the feature envy from the
Authorization class:

public class Authorization

{
private ProblemCode.Authentication _authentication;

public Authorization(ProblemCode.Authentication authentication)

{
_authentication = authentication;
}

public bool IsAuthorized(string role)

{
return _authentication.IsAuthenticated() &&
role.Contains("Administrator");
}
}

You will see that the Authorization class is a lot smaller now, and only does what it
needs to. There is no longer any feature envy.

Next up, we will look at an inappropriate intimacy code smell.

Inappropriate intimacy
A class engages in inappropriate intimacy when it relies on the implementation details held
in a separate class. Does the class that has this reliance really need to exist? Can it be
merged with the class that it relies on? Or is there shared functionality that is better off
being extracted into its own class?

Classes should not rely on each other as this causes coupling, and it can also affect
cohesion. A class should ideally be self-contained. And classes should really know as little
about each other as possible.

[ 407 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

Indecent exposure
When a class reveals its internal details, this is known as indecent exposure. This breaks
the OOP principle of encapsulation. Only that which should be public should be public. All
other implementations that don't need to be public should be hidden by using the
appropriate access modifiers.

Data values should not be public. They should be private, and they should only be
modifiable via constructors, methods, and properties. And they should only be retrievable
via properties.

The large class (aka the God object)

The large class, also known as the God object, is all things to all parts of the system. It is a
large, unwieldy class that simply does far too much. When you attempt to read the object,
the intent of the code may be clear when you read the class name and see what namespace
it is in, but then when you come to look at the code, the intent of the code can become lost.

A well-written class should have the name of its intent and should be placed in the
appropriate namespace. The contents of the class should follow the company coding
standards. Methods should be kept as small as possible, and method parameters should be
kept to the absolute bare minimum. Only the methods that belong in the class should be in
the class. Member variables, properties, and methods that don't belong in the class should
be removed and placed in the correct files in the correct namespace.

To keep classes small and focused, don't inherit classes if there is no need. If there is a class
that has five methods, and you will only ever use one of them, is it possible to move that
method out into its own reusable class? Remember the single responsibility principle. A
class should only have a single responsibility. For example, a file class should only handle
operations and behaviors associated with files. A file class should not be performing
database operations. You get the idea.

When writing a class, your aim is to make it as small, clean, and readable as you can.

The lazy class (aka the freeloader and the lazy

object)
A freeloading class is one that hardly does anything to be useful. When you encounter
such classes, you can merge their contents with other classes that have the same kind of
intentions.

[ 408 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

You can also attempt to collapse the inheritance hierarchy. Remember that the ideal depth
of inheritance is 1. And so, if your classes have a larger value for their depth of inheritance,
then they are good candidates for moving back up the inheritance tree. You may also want
to consider using inline classes for really small classes.

The middleman class

The middleman class does no more than delegate functionality to other objects. In
situations like this, you can get rid of the middleman and deal with the objects that carry
out the responsibility directly.

Also, remember that you need to keep the depth of inheritance down. So if you cannot get
rid of the class, look to merge it with existing classes. Look at the overall design of that area
of code. Could it all be refactored in some way to reduce the amount of code and the
number of different classes?

The orphan class of variables and constants

It is not really good practice to have a lone class that holds variables and constants for
multiple different parts of the application. When you encounter such a situation, it can be
hard for the variables to have any real meaning and their context can be lost. It is better to
move constants and variables to areas that use them. If constants and variables will be used
by multiple classes, then they should be assigned to a file within the root of the namespace
they will be used in.

Primitive obsession
Source code that uses primitive values rather than objects for certain tasks such as range
values and formatted strings such as credit cards, postcodes, and phone numbers suffers
from primitive obsession. Other signs include constants used for field names, and
information stored inappropriately stored in constants.

[ 409 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

Refused bequest
When a class inherits from another class but does not use all its methods, then this is known
as refused bequest. A common reason for this happening is when the subclass is
completely different from the base class. For example, a building base class is used by
different building types, but then a car object inherits building because it has properties
and methods to do with windows and doors. This is clearly wrong.

When you encounter this, consider whether a base class is necessary. If it is, then create one
and then inherit from it. Otherwise, add the functionality to the class that was inherited
from the wrong type.

Speculative generality
A class that is programmed with functionality that is not needed now but may be needed in
the future is suffering from speculative generality. Such code is dead code and adds
maintenance overhead as well as code bloat. It is best to remove these classes when you see
them.

Tell, Don't Ask

The Tell, Don't Ask software principle informs us as programmers that we are to bundle
data with the methods that will operate on that data. Our objects must not ask for data and
then operate on it! They must tell the logic of an object to perform a specific task on that
object's data.

If you find objects that contain logic and that ask other objects for data to carry out their
operations, then combine the logic and the data into a single class.

Temporary fields
Temporary fields are member variables that are not needed for an object's entire lifetime.

You can perform refactoring by removing the temporary fields and the methods that
operate upon them to their own class. You will end up with clearer code that is well
organized.

[ 410 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

Method-level smells
Method-level code smells are problems within the method itself. Methods are the work-
horses that either make software function well or poorly. They should be well organized
and do only what they are expected to do—no more and no less. It is important to know the
kinds of problems and issues that can arise because of poorly constructed methods. We will
address what to look out for in terms of method-level code smells, and what we can do to
address them. We'll start with the black sheep method first.

The black sheep method

Out of all the methods in the class, a black sheep method will be noticeably different. When
you encounter a black sheep method, you must consider the method objectively. What is its
name? What is the method's intent? When you have answered these questions, then you
can decide to remove the method and place it where it truly belongs.

Cyclomatic complexity
When a method has too many loops and branches, this is known as cyclomatic complexity.
This code smell is also a class-level code smell, and we have already seen how we can
reduce the problems with branching when we looked at replacing switch and if
statements. As for loops, they can be replaced with LINQ statements. LINQ statements
have the added benefit of being a functional code since LINQ is a functional query
language.

Contrived complexity
When a method is unnecessarily complex and can be simplified, this complexity is termed
contrived complexity. Simplify the method to make sure that its contents are human-
readable and understandable. Then, look to refactor the method and reduce the size to the
smallest number of lines that is practical.

Dead code
When a method exists but is not used, this is known as dead code. The same goes for
constructors, properties, parameters, and variables. They should be identified and
removed.

[ 411 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

Excessive data return

When a method returns more data than is needed by each client that calls it, this code smell
is known as excessive data return. Only the data that is required should be returned. If you
find that there are groups of objects with different requirements, then you should maybe
consider writing different methods that appeal to both groups and only return what is
necessary to those groups.

Feature envy
A method that has feature envy spends more time accessing data in other objects than it
does in its own object. We have already seen this in action when we looked at feature envy
under class-level code smells.

A method should be kept small, and most of all, its main functionality should be localized
to that method. If it is doing more in other methods than its own, then there is scope for
moving some of the code out of the method and into its own method.

Identifier size
Identifiers can be either too short or too long. Identifiers should be descriptive and succinct.
The main thing to consider when naming variables is the context and location. In a
localized loop, a single letter may be appropriate. But if the identifier is at the class level,
then it will need a human-understandable name to give it context. Avoid using names that
lack context, and that are ambiguous or cause confusion.

Inappropriate intimacy
Methods that rely too heavily on implementation details in other methods or classes
display inappropriate intimacy. These methods need to be refactored and possibly even
removed. The main thing to bear in mind is that the methods use the internal fields and
methods of another class.

To perform refactoring, you can move the methods and fields to where they actually need
to be used. Alternatively, you can extract the fields and methods into a class of their own.
Inheritance can replace delegation when the subclass is being intimate with the superclass.

[ 412 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

Long lines (aka God lines)

Long lines of code can be very hard to read and decipher. This makes it difficult for
programmers to debug and refactor such code. Where it is possible, the line can be
formatted so that any periods and any code after a comma appears on a new line. But such
code should also be refactored to make it small.

Lazy methods
A lazy method is one that does very little work. It may delegate its work to other methods,
and it may simply call a method on another class that does what it is supposed to. If any of
these are the case, then it may pay to get rid of the methods and place code within the
methods where it is needed. You could, for instance, use an inline function such as a
lambda.

Long methods (aka God methods)

A long method is one that has outgrown itself. Such methods may lose their intent and
perform more tasks than they are expected to. You can use the IDE to select parts of the
method, and then select extract method or extract class to move portions of the method to
their own method and even their own class. A method should only be responsible for doing
a single task.

Long parameter lists (aka too many parameters)

Three or more parameters are classed as the long parameter list code smell. You can tackle
this problem by replacing the parameters with a method call. An alternative is to replace
the parameters with a parameter object.

Message chains
A message chain occurs when a method calls an object that calls another object that calls
another object and so on. Previously, you saw how to deal with message chains when we
looked at the Law of Demeter. Message chains break this law, as a class should only
communicate with its nearest neighbor. Refactor the classes to move the required state and
behavior closer to where it is needed.

[ 413 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

The middleman method

When all a method does is delegate work out to others to complete, it is a middleman
method and can be refactored and removed. But if there is functionality that can't be
removed, then merge it in the area that it is being used in.

Oddball solutions
When you see multiple methods doing the same thing but doing it differently, then this is
an oddball solution. Choose the method that best implements the task, and then replace the
method calls to the other methods with calls to the best method. Then, delete the other
methods. This will leave only one method and one way of implementing the task that can
be reused.

Speculative generality
A method that is not used anywhere in the code is known as a speculative generality code
smell. It is essentially dead code, and all dead code should be removed from the system.
Such code provides a maintenance overhead and also provides unnecessary code bloat.

Summary
In this chapter, you have been introduced to a variety of code smells and how to remove
them through refactoring. We have stated that there are application-level code smells that
permeate throughout all the layers of the application, class-level code smells that run
throughout the class, and method-level code smells that affect the individual methods.

First of all, we covered the application-level code smells, which consisted of Boolean
blindness, combinatorial explosion, contrived complexity, data clump, deodorant
comments, duplicate code, lost intent, mutation of variables, oddball solutions, shotgun
surgery, solution sprawl, and uncontrolled side effects.

We then went on to look at class-level code smells, including cyclomatic complexity,

divergent change, downcasting, excessive literal use, feature envy, inappropriate intimacy,
indecent exposure, and the large object, also known as the God object. We also covered the
lazy class, also known as the freeloader and the lazy object; middleman; orphan classes of
variables and constants; primitive obsession; refused bequest; speculative generality; Tell,
Don't Ask; and temporary fields.

[ 414 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

Finally, we moved on to method-level code smells. We discussed black sheep; cyclomatic

complexity; contrived complexity; dead code; feature envy; identifier size; inappropriate
intimacy; long lines, also known as the God lines; the lazy method; the long method, also
known as the God method; the long parameter list, also known as too many parameters;
message chains; middleman; oddball solutions; and speculative generality.

In the next chapter, we will be continuing our look at code refactoring with the use of
ReSharper.

Questions
1. What are the three main categories of code smell?
2. Name the different types of application-level code smells.
3. Name the different types of class-level code smells.
4. Name the different types of method-level code smells.
5. What kinds of refactoring are you able to perform in order to clean up various
code smells?
6. What is cyclomatic complexity?
7. How can we overcome cyclomatic complexity?
8. What is contrived complexity?
9. How can we overcome contrived complexity?
10. What is a combinatorial explosion?
11. How do we overcome a combinatorial explosion?
12. What should you do when you find deodorant comments?
13. If you have bad code but don't know how to fix it, what should you do?
14. What is a good place to ask questions and get answers when it comes to
programming issues?
15. In what ways can a long parameter list be reduced?
16. How can a large method be refactored?
17. What is the maximum length for a clean method?
18. Within what range of numbers should your program's cyclomatic complexity be?
19. What is the ideal depth of inheritance value?
20. What is speculative generality and what should you do about it?
21. If you encounter an oddball solution, what course of action should you take?
22. What refactorings would you perform if you encountered a temporary field?

[ 415 ]
Refactoring C# Code – Identifying Code Smells Chapter 13

23. What is a data clump, and what should you do about it?
24. Explain the refused bequest code smell.
25. What law do message chains break?
26. How should message chains be refactored?
27. What is feature envy?
28. How do you remove feature envy?
29. What pattern can you use to replace switch statements that return objects?
30. How can we replace if statements that return objects?
31. What is solution sprawl, and what can be done to tackle it?
32. Explain the Tell, don't ask! principle.
33. How does the Tell, don't ask! principle get broken?
34. What are the symptoms of shotgun surgery, and how should they be addressed?
35. Explain lost intent and what can be done about it.
36. How can loops be refactored, and what benefits do the refactorings bring?
37. What is a divergent change, and how would you go about refactoring it?

Further reading
Refactoring - Improving the Design of Existing Code by Martin Fowler and Kent
Beck.
https://refactoring.guru/refactoring: A good site on design patterns and
code smells.
https://www.dofactory.com/net/design-patterns: A very good C#-based site
on various design patterns.

[ 416 ]
14
Refactoring C# Code –
Implementing Design Patterns
Half the battle in programming clean code is in the correct implementation and usage of
design patterns. Design patterns themselves can become code smells. A design pattern
becomes a code smell when it is used to over-engineer something that is rather simple to
implement.

You have already seen the use of design patterns in writing clean code and refactoring code
smells in the previous chapters of this book. Specifically, we have implemented the adapter
pattern, the decorator pattern, and the proxy pattern. These patterns were implemented in
the right way to accomplish the task at hand. They were kept simple and they most
certainly did not complicate the code. So, when used for their proper purpose, design
patterns are really useful in removing code smells, thus leaving your code nice, clean, and
fresh.

In this chapter, we will address the Gang of Four (GoF) creational, structural, and
behavioral design patterns. Design patterns are not set in stone and you don't have to be
strict in their implementation. But having code samples can help you transition
from just having head knowledge to having the practical skills needed to correctly
implement and use design patterns.

In this chapter, we will be covering the following topics:

Implementing creational design patterns

Implementing structural design patterns
Overview of behavioral design patterns
Refactoring C# Code – Implementing Design Patterns Chapter 14

By the end of this chapter, you will have the following skills:

The ability to understand, describe, and program different creational design

patterns
The ability to understand, describe, and program different structural design
patterns
An understanding of an overview of behavioral design patterns

We will begin our overview of GoF design patterns by addressing creational design
patterns.

Technical requirements
Visual Studio 2019
A Visual Studio 2019 .NET Framework console application as your working
project
The complete source code for this chapter: https://github.com/
PacktPublishing/Clean-Code-in-C-/tree/master/CH14/CH14_DesignPatterns

Implementing creational design patterns

From a programmer's perspective, we use creational design patterns when we perform
object creation. Patterns are selected based on the task at hand. There are five creational
design patterns:

Singleton: The singleton pattern ensures that only one instance of an object will
exist at the application level.
Factory method: A factory pattern is used to create objects without using the
class to be used.
Abstract factory: Without the specification of their concrete classes, groups of
related or dependent objects are instantiated by the abstract factory.
Prototype: Specifies the type of prototype to create, and then creates copies of the
prototype.
Builder: Separates object construction from its representation.

We will now begin implementing each of these patterns, starting with the singleton design
pattern.

[ 418 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

Implementing the singleton pattern

The singleton design pattern only allows one instance of a class with global access to it. Use
the singleton pattern when all operations within a system must be coordinated by exactly
one object:

The participant in this pattern is singleton—a class that is responsible for managing its own
instance and ensures that there is only one instance of itself running in the entire system.

We are now going to implement the singleton design pattern:

1. Add a folder called Singleton to the CreationalDesignPatterns folder.

Then, add a class called Singleton:
public class Singleton {
private static Singleton _instance;

protected Singleton() { }

public static Singleton Instance() {

return _instance ?? (_instance = new Singleton());
}
}

2. The Singleton class stores a static copy of an instance of itself. You cannot
instantiate the class because the constructor is marked as protected. The
Instance() method is static. It checks to see whether an instance of the
Singleton class exists. If it does, then it is returned. If it does not exist, then the
instance is created and returned. Now, we'll add the code to call it:
var instance1 = Singleton.Instance();
var instance2 = Singleton.Instance();

if (instance1.Equals(instance2))
Console.WriteLine("Instance 1 and instance 2 are the same
instance of Singleton.");

[ 419 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

3. We declare two instances of the Singleton class, and then compare them to see
whether they are the same instance. You can see the output in the following
screenshot:

As you can see, we have a working class that implements the singleton design pattern. Next
up, we'll tackle the factory method design pattern.

Implementing the factory method pattern

The factory method design pattern creates objects that let their subclasses implement their
own object creation logic. Use this design pattern when you want to keep object
instantiation in a single place and need to generate a specific group of related objects:

The participants in this project are as follows:

Product: The abstract product created by the factory method

ConcreteProduct: Inherits the abstract product
Creator: An abstract class with an abstract factory method
Concrete Creator: Inherits the abstract creator and overrides the factory
method

[ 420 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

We will now implement the factory method:

1. Add a folder to the CreationalDesignPatterns folder

called FactoryMethod. Then, add the Product class:
public abstract class Product {}

2. The Product class defines the objects that are created by the factory method.
Add the ConcreteProduct class:
public class ConcreteProduct : Product {}

3. The ConcreteProduct class inherits the Product class. Add the Creator class:
public abstract class Creator {
public abstract Product FactoryMethod();
}

4. The Creator class will be inherited by the ConcreteFactory class, which will
implement FactoryMethod(). Add the ConcreteCreator class:
public class ConcreteCreator : Creator {
public override Product FactoryMethod() {
return new ConcreteProduct();
}
}

5. The ConcreteCreator class inherits the Creator class and overrides

the FactoryMethod(). A new ConcreteProduct class is returned by the
method. The following code demonstrates the factory method in use:
var creator = new ConcreteCreator();
var product = creator.FactoryMethod();
Console.WriteLine($"Product Type: {product.GetType().Name}");

We have created a new instance of the ConcreteCreator class. Then, we called

the FactoryMethod() to create a new product. The name of the product created by the
factory method is then output to the console window, as shown:

[ 421 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

Now that we know how to implement the factory method design pattern, we will move on
to implementing the abstract factory design pattern.

Implementing the abstract factory pattern

Without the specification of their concrete classes, groups of related or
dependent objects, referred to as families, are instantiated using the abstract factory design
pattern:

The participants in this pattern are as follows:

AbstractFactory: The abstract factory, which is implemented by concrete

factories
ConcreteFactory: Creates concrete products
AbstractProduct: The abstract product that concrete products will inherit
Product: Inherits AbstractProduct and is created by the concrete factory

[ 422 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

We will now start implementing the pattern:

1. Add a folder to the project called CreationalDesignPatterns.

2. Add a folder to the CreationalDesignPatterns folder
called AbstractFactory.
3. In the AbstractFactory folder, add the AbstractFactory class:
public abstract class AbstractFactory {
public abstract AbstractProductA CreateProductA();
public abstract AbstractProductB CreateProductB();
}

4. AbstractFactory contains two abstract methods for creating abstract products.

Add the AbstractProductA class:
public abstract class AbstractProductA {
public abstract void Operation(AbstractProductB productB);
}

5. The AbstractProductA class has a single abstract method, which performs an

operation on AbstractProductB. Now, add the AbstractProductB class:
public abstract class AbstractProductB {
public abstract void Operation(AbstractProductA productA);
}

6. The AbstractProductB class has a single abstract method, which performs an

operation on AbstractProductA. Add the ProductA class:
public class ProductA : AbstractProductA {
public override void Operation(AbstractProductB productB) {
Console.WriteLine("ProductA.Operation(ProductB)");
}
}

7. ProductA inherits AbstractProductA and overrides

the Operation() method, which interacts with AbstractProductB.
The Operation() method in this example prints out a console message. Do the
same for the ProductB class:
public class ProductB : AbstractProductB {
public override void Operation(AbstractProductA productA) {
Console.WriteLine("ProductB.Operation(ProductA)");
}
}

[ 423 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

8. ProductB inherits AbstractProductB and overrides

the Operation() method, which interacts with AbstractProductA.
The Operation() method in this example prints out a console message. Add
the ConcreteFactory class:
public class ConcreteProduct : AbstractFactory {
public override AbstractProductA CreateProductA() {
return new ProductA();
}

public override AbstractProductB CreateProductB() {

return new ProductB();
}
}

9. ConcreteFactory inherits the AbstractFactory class and overrides the two

product creation methods. Each method returns a concrete class. Add
the Client class:
public class Client
{
private readonly AbstractProductA _abstractProductA;
private readonly AbstractProductB _abstractProductB;

public Client(AbstractFactory factory) {

_abstractProductA = factory.CreateProductA();
_abstractProductB = factory.CreateProductB();
}

public void Run() {

_abstractProductA.Operation(_abstractProductB);
_abstractProductB.Operation(_abstractProductA);
}
}

10. The Client class declares two abstract products. Its constructor takes
an AbstractFactory class. Inside the constructor, both declared abstract
products are assigned their respective concrete products by the factory.
The Run() method executes Operation() on both products. The following code
executes our abstract factory example:
AbstractFactory factory = new ConcreteProduct();
Client client = new Client(factory);
client.Run();

11. Run the code and you will see the following output:

[ 424 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

A good reference implementation of the abstract factory is the ADO.NET

2.0 DbProviderFactory abstract class. An article called Abstract Factory
Design Pattern in ADO.NET 2.0 by Moses Soliman on C# Corner is a nice
write-up on DbProviderFactory about the implementation of the
abstract factory design pattern. Here is the link:

https://www.c-sharpcorner.com/article/abstract-factory-design-
pattern-in-ado-net-2-0/.

We have successfully implemented the abstract factory design pattern. Now, we will
implement the prototype pattern.

Implementing the prototype pattern

The prototype design pattern is used to create an instance of a prototype, and then to create
new objects by cloning the prototype. Use this pattern when the cost of creating objects
directly is expensive. With this pattern, you can cache the object and return a clone when
needed:

[ 425 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

The participants in the prototype design pattern are as follows:

Prototype: An abstract class that provides a method for cloning itself

ConcretePrototype: Inherits the prototype and overrides the Clone() method
to return a memberwise clone of the prototype
Client: Requests new clones of the prototype

We will now implement the prototype design pattern:

1. Add a folder called Prototype to the CreationalDesignPatterns folder, and

then add the Prototype class:
public abstract class Prototype {
public string Id { get; private set; }

public Prototype(string id) {

Id = id;
}

public abstract Prototype Clone();

}

2. Our Prototype class must be inherited. Its constructor requires an identifying

string to be passed in that is stored at the class level. A Clone() method is
provided, which the subclass will override. Now, add the ConcretePrototype
class:
public class ConcretePrototype : Prototype {
public ConcretePrototype(string id) : base(id) { }

public override Prototype Clone() {

return (Prototype) this.MemberwiseClone();
}
}

3. The ConcretePrototype class inherits from the Prototype class. Its

constructor takes an identifying string and passes that string into the constructor
of the base class. It then overrides the clone method to provide a shallow copy of
the current object by calling the MemberwiseClone() method and returning the
clone that is cast to the type of Prototype. Now for the code that demonstrates
the prototype design pattern in use:
var prototype = new ConcretePrototype("Clone 1");
var clone = (ConcretePrototype)prototype.Clone();
Console.WriteLine($"Clone Id: {clone.Id}");

[ 426 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

Our code creates a new instance of the ConcretePrototype class with an identifier of
"Clone 1". We then clone the prototype and cast it to the ConcretePrototype type.
Then, we print the clone's identifier to the console window, as shown:

As we can see, the clone has the same identifier as the prototype that it was cloned from.

For a very detailed article of a real-world example, refer to an excellent

article called Prototype Design Pattern with Real-World Scenario, by Akshay
Patel, on C# Corner. Here is the link: https://www.c-sharpcorner.com/
UploadFile/db2972/prototype-design-pattern-with-real-world-
scenario624/.

We will now implement our final creational design pattern called the builder design
pattern.

Implementing the builder pattern

The builder design pattern separates the object's construction from its representation. As a
result, you can use the same construction method to create different representations of the
object. Use the builder design pattern when you have a complex object that needs to be
built up and connected in stages:

[ 427 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

The participants in the builder design pattern are as follows:

Director: A class that receives a builder via its constructor, and then calls each
of the build methods on the builder object
Builder: An abstract class that provides abstract build methods and an abstract
method for returning the built object
ConcreteBuilder: A concrete class that inherits the Builder class, overrides
the builder methods to actually build the object, and overrides the result method
to return the fully built object

Let's start implementing our final creational design pattern—the builder design pattern:

1. Start by adding a folder called Builder to the CreationalDesignPatterns

folder. Then, add the Product class:
public class Product {
private List<string> _parts;

public Product() {
_parts = new List<string>();
}

public void Add(string part) {

_parts.Add(part);
}

public void PrintPartsList() {

var sb = new StringBuilder();
sb.AppendLine("Parts Listing:");
foreach (var part in _parts)
sb.AppendLine($"- {part}");
Console.WriteLine(sb.ToString());
}
}

2. In our example, the Product class keeps a list of parts. These parts are strings.
The list is initialized in the constructor. Parts are added by the Add() method,
and when our object is fully constructed, we can call the PrintPartsList()
method to print the list of parts that make up the object to the console window.
Now, add the Builder class:
public abstract class Builder
{
public abstract void BuildSection1();
public abstract void BuildSection2();

[ 428 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

public abstract Product GetProduct();

}

3. Our Builder class will be inherited by concrete classes that will override its
abstract methods to build the object and return it. We'll now add the
ConcreteBuilder class:

public class ConcreteBuilder : Builder {

private Product _product;

public ConcreteBuilder() {
_product = new Product();
}

public override void BuildSection1() {

_product.Add("Section 1");
}

public override void BuildSection2() {

_product.Add(("Section 2"));
}

public override Product GetProduct() {

return _product;
}
}

4. Our ConcreteBuilder class inherits the Builder class. The class stores the
instance of the object to be constructed. The build methods are overridden and
parts are added to the product via the product's Add() method. The product is
returned to the client via the GetProduct() method call. Add the Director
class:
public class Director
{
public void Build(Builder builder)
{
builder.BuildSection1();
builder.BuildSection2();
}
}

[ 429 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

5. The Director class is a concrete class that takes a Builder object via its
Build() method and calls the build methods on the Builder object to build the
object. All we need now is the code to demonstrate the builder design pattern in
action:
var director = new Director();
var builder = new ConcreteBuilder();
director.Build(builder);
var product = builder.GetProduct();
product.PrintPartsList();

6. We create a director and builder. Then, the director builds the product. The
product is then assigned, and its parts list is printed out to the console window,
as shown:

Everything is working as it should be.

In .NET Framework, the System.Text.StringBuilder class is an example of the builder

design pattern in the real world. Using string concatenation with the plus (+) operator is
slower than using the StringBuilder class when concatenating five or more lines. String
concatenation with the + operator is faster than StringBuilder when you have less than
five concatenation lines, but slower when you have more than five lines to concatenate. The
reason for this is that each time you create a string with the + operator, you are recreating
the string since strings are immutable on the heap. But StringBuilder allocates buffer
space on the heap. Then, characters are written to the buffer space. For only a small number
of lines, the + operator is faster because of the overhead of creating the buffer when using
the string builder. But when there are more than five lines, there is a noticeable difference
when using StringBuilder. In big data projects where there may be hundreds of
thousands or even millions of string concatenations taking place, the string concatenation
strategy that you decide to employ will either perform fast or sluggishly. Let's create a
simple demonstration. Create a new class called StringConcatenation, and then add the
following code:
private static DateTime _startTime;
private static long _durationPlus;
private static long _durationSb;

[ 430 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

The _startTime variable holds the current start time of the method execution. The
_durationPlus variable holds the duration of the method execution as the number of
ticks when using the + operator to concatenate, and _durationSb holds the duration of the
operation as the number of ticks for the StringBuilder concatenation. Add the
UsingThePlusOperator() method to the class:

public static void UsingThePlusOperator()

{
_startTime = DateTime.Now;
var text = string.Empty;
for (var x = 1; x <= 10000; x++)
{
text += $"Line: {x}, I must not be a lazy programmer, and should
continually develop myself!\n";
}
_durationPlus = (DateTime.Now - _startTime).Ticks;
Console.WriteLine($"Duration (Ticks) Using Plus Operator:
{_durationPlus}");
}

The UsingThePlusOperator() method demonstrates the time taken when concatenating

10,000 strings using the + operator. The time taken to process the string concatenation is
stored as the number of ticks fired. There are 10,000 ticks per millisecond. Now, add the
UsingTheStringBuilder() method:

public static void UsingTheStringBuilder()

{
_startTime = DateTime.Now;
var sb = new StringBuilder();
for (var x = 1; x <= 10000; x++)
{
sb.AppendLine(
$"Line: {x}, I must not be a lazy programmer, and should
continually develop myself!"
);
}
_durationSb = (DateTime.Now - _startTime).Ticks;
Console.WriteLine($"Duration (Ticks) Using StringBuilder:
{_durationSb}");
}

[ 431 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

This method is the same as the previous one, except we perform string concatenation using
the StringBuilder class. We'll now add the code to print out the time difference, called
PrintTimeDifference():

public static void PrintTimeDifference()

{
var difference = _durationPlus - _durationSb;
Console.WriteLine($"That's a time difference of {difference} ticks.");
Console.WriteLine($"{difference} ticks =
{TimeSpan.FromTicks(difference)} seconds.\n\n");
}

The PrintTimeDifference() method calculates the time difference by subtracting the

StringBuilder ticks from the + ticks. The difference in ticks is then printed to the console,
followed by a line that translates the ticks into seconds. Here is the code to test our methods
so that we can see the time difference in the two concatenation methods:
StringConcatenation.UsingThePlusOperator();
StringConcatenation.UsingTheStringBuilder();
StringConcatenation.PrintTimeDifference();

When you run the code, you will see the times and time difference in the console window,
as shown:

As you can see from the screenshot, StringBuilder is much faster. With small amounts of
data, you don't really see a difference with the naked eye. But the difference is noticeable to
the naked eye when the data lines being processed greatly increase in number.

Another example that comes to mind for using the builder pattern is report construction. If
you consider banded reports, the bands are essentially sections that need to be built up
from various sources. So, you could have the main part, and then each subreport as a
different part. The final report would be the amalgamation of these various parts. So, you
could have code like the following to build a report:
var report = new Report();
report.AddHeader();
report.AddLastYearsSalesTotalsForAllRegions();

[ 432 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

report.AddLastYearsSalesTotalsByRegion();
report.AddFooter();
report.GenerateOutput();

Here, we are creating a new report. We start by adding the header. Then, we add last year's
sales figures, combined for all regions, followed by last year's sales figures, broken down by
region. We then add a footer to the report and complete the process by generating the
report output.

So, you've seen the default implementation of the builder pattern from the UML diagram.
Then, you implemented string concatenation using the StringBuilder class, which helps
you build strings in a performant manner. Finally, you learned how the builder pattern can
be useful in building up the sections of a report and generating its output.

Well, that concludes our implementations of the creational design patterns. We will now
move on to implementing some structural design patterns.

Implementing structural design patterns

As programmers, we use structural patterns to improve the overall structure of our code.
So, when code is encountered that lacks structure and is not at its cleanest, we can use the
patterns mentioned in this section to restructure the code and make it clean. There are
seven structural design patterns:

Adapter: Use this pattern to enable classes with incompatible interfaces to work
cleanly together.
Bridge: Use this pattern to loosely couple code by decoupling an abstraction
from its implementation.
Composite: Use this pattern to aggregate objects and provide a uniform way of
working with individual and object compositions.
Decorator: Use this pattern to keep the interface the same while dynamically
adding new functionality to the object.
Façade: Use this pattern to simplify larger and more complex interfaces.
Flyweight: Use this pattern to conserve memory and pass shared data between
objects.
Proxy: Use this pattern between a client and an API to intercept calls between the
client and the API.

[ 433 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

We have already touched on the adapter, decorator, and proxy patterns in previous
chapters, so they won't be covered again in this chapter. Now, we'll start implementing our
structural design patterns, starting with the bridge pattern.

Implementing the bridge pattern

We use the bridge pattern to decouple abstractions from their implementations so that they
are not bound at compile time. Both the abstraction and implementation can vary without
impacting the client.

Use the bridge design pattern if you require runtime binding of the implementation or
sharing of the implementation between multiple objects, if a number of classes exist as a
result of interface coupling and various implementations, or if there is a need for
orthogonal class hierarchies to be mapped:

The participants of the bridge design pattern are as follows:

Abstraction: An abstract class that contains abstract operations

RefinedAbstraction: Inherits the Abstraction class and overrides the
Operation() method

[ 434 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

Implementor: An abstract class with an abstract Operation() method

ConcreteImplementor: Inherits the Implementor class and overrides the
Operation() method

We will now implement the bridge design pattern:

1. Start by adding the StructuralDesignPatterns folder to the project, and then

in that folder, add the Bridge folder. Then, add the Implementor class:
public abstract class Implementor {
public abstract void Operation();
}

2. The Implementor class has just a single abstract method, called Operation().
Add the Abstraction class:
public class Abstraction {
protected Implementor implementor;

public Implementor Implementor {

set => implementor = value;
}

public virtual void Operation() {

implementor.Operation();
}
}

3. The Abstraction class has a protected field that holds the Implementor object,
which is set via the Implementor property. A virtual method called
Operation() calls the Operation() method on the implementor. Add the
RefinedAbstraction class:

public class RefinedAbstraction : Abstraction {

public override void Operation() {
implementor.Operation();
}
}

[ 435 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

4. The RefinedAbstraction class inherits the Abstraction class and overrides

the Operation() method to call the Operation() method on the implementor.
Now, add the ConcreteImplementor class:
public class ConcreteImplementor : Implementor {
public override void Operation() {
Console.WriteLine("Concrete operation executed.");
}
}

5. The ConcreteImplementor class inherits the Implementor class and overrides

the Operation() method to print out a message to the console. The code to run
the bridge design pattern example is as follows:
var abstraction = new RefinedAbstraction();
abstraction.Implementor = new ConcreteImplementor();
abstraction.Operation();

We create a new RefinedAbstraction instance and then set its implementor to a new
instance of ConcreteImplementor. Then, we call the Operation() method. The output
from our example bridge implementation is as follows:

As you can see, we successfully executed the concrete operation in the concrete
implementor class. The next pattern we will look at is the composite design pattern.

Implementing the composite pattern

With the composite design pattern, objects are composed of tree structures to represent
part-whole hierarchies. This pattern enables you to treat individual objects and
compositions of objects in a uniform manner.

Use this pattern when you need to ignore the differences between individual objects and
object compositions, when you need tree structures to represent hierarchies, and when a
hierarchical structure requires generic functionality across its structure:

[ 436 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

The participants in the composite design pattern are as follows:

Component: Composed objects interface

Leaf: A leaf in the composition that has no children
Composite: Stores child components and performs operations
Client: Manipulates compositions and leaves via the component interface

It's time to implement the composite pattern:

1. Add a new folder called Composite to the StructuralDesignPatterns class.

Then, add the IComponent interface:
public interface IComponent {
void PrintName();
}

2. The IComponent interface has a single method, which will be implemented by

both leaves and composites. Add the Leaf class:
public class Leaf : IComponent {
private readonly string _name;

public Leaf(string name) {

_name = name;

[ 437 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

public void PrintName() {

Console.WriteLine($"Leaf Name: {_name}");
}
}

3. The Leaf class implements the IComponent interface. Its constructor takes a
name and stores it, and the PrintName() method prints the name of the leaf to
the console window. Add the Composite class:
public class Composite : IComponent {
private readonly string _name;
private readonly List<IComponent> _components;

public Composite(string name) {

_name = name;
_components = new List<IComponent>();
}

public void Add(IComponent component) {

_components.Add(component);
}

public void PrintName() {

Console.WriteLine($"Composite Name: {_name}");
foreach (var component in _components) {
component.PrintName();
}
}
}

4. The Composite class implements the IComponent interface in the same way that
the leaf does. Additionally, Composite stores a list of components that are added
via the Add() method. Its PrintName() method prints out its own name,
followed by the names of each of the components in the list. Now, we'll add the
code to test our composite design pattern implementation:
var root = new Composite("Classification of Animals");
var invertebrates = new Composite("+ Invertebrates");
var vertebrates = new Composite("+ Vertebrates");

var warmBlooded = new Leaf("-- Warm-Blooded");

var coldBlooded = new Leaf("-- Cold-Blooded");
var withJointedLegs = new Leaf("-- With Jointed-Legs");
var withoutLegs = new Leaf("-- Without Legs");

[ 438 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

invertebrates.Add(withJointedLegs);
invertebrates.Add(withoutLegs);

vertebrates.Add(warmBlooded);
vertebrates.Add(coldBlooded);

root.Add(invertebrates);
root.Add(vertebrates);

root.PrintName();

5. As you can see, we create our composites and then our leaves. We then add the
leaves to the appropriate composites. Then, we add our composites to the root
composite. Finally, we call the root composite's PrintName() method, which
will print the root's name, along with the names of all the components and leaves
in the hierarchy. You can see the output, as follows:

Our composite implementation is working as expected. The next pattern we will implement
is the façade design pattern.

Implementing the façade pattern

The façade pattern is designed to make using API subsystems easier to use. Use this pattern
to hide a large and complex system behind a much simpler interface for your clients to use.
The main reason that programmers will implement this pattern is that the system they are
having to use or work on is too complex and very hard to understand.

[ 439 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

Other reasons why this pattern is employed include if too many classes are dependent on
one another, or simply because programmers don't have access to the source code:

The participants in the façade pattern are as follows:

Facade: The simple interface, which acts as a go-between between the client and a
more complex system of subsystems
Subsystem Classes: The subsystem classes, which are directly removed from
client access and are directly accessed by the façade

We are now going to implement the façade design pattern:

1. Add a folder called Facade to the StructuralDesignPatterns folder. Then,

add the SubsystemOne and SubsystemTwo classes:
public class SubsystemOne {
public void PrintName() {
Console.WriteLine("SubsystemOne.PrintName()");
}
}

public class SubsystemOne {

public void PrintName() {
Console.WriteLine("SubsystemOne.PrintName()");
}
}

[ 440 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

2. These classes have a single method that prints the class name and method name
to the console window. Now, let's add the Facade class:
public class Facade {
private SubsystemOne _subsystemOne = new SubsystemOne();
private SubsystemTwo _subsystemTwo = new SubsystemTwo();

public void SubsystemOneDoWork() {

_subsystemOne.PrintName();
}

public void SubsystemTwoDoWork() {

_subsystemTwo.PrintName();
}
}

3. The Facade class creates member variables for each system that it has
knowledge of. It then provides a series of methods that will access various
portions of each of the subsystems when requested to do so. We will add the
code to test our implementation:
var facade = new Facade();
facade.SubsystemOneDoWork();
facade.SubsystemTwoDoWork();

4. All we have to do is create a Facade variable, and then we can call the methods
that execute method calls in the subsystems. You should see the following
output:

Time to look at our final structural pattern called the flyweight pattern.

[ 441 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

Implementing the flyweight pattern

The flyweight design pattern is used to efficiently process a large number of fine-grained
objects by reducing the overall object count. Use this pattern to increase performance and
reduce the memory footprint by reducing the number of objects that you create:

The participants in the flyweight design pattern are as follows:

Flyweight: Provides an interface for flyweights so that they can receive an

extrinsic state and act on it
ConcreteFlyweight: A sharable object that adds storage for the intrinsic state
UnsharedConcreteFlyweight: Used when flyweights don't need to be shared
FlyweightFactory: Correctly manages flyweight objects and shares
them properly
Client: Maintains flyweight references and computes or stores the extrinsic
state of flyweights

[ 442 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

Extrinsic state means that it is not part of the essential nature of the object
and that it originates externally to the object. Intrinsic state means that the
state belongs to the object and is essential to the object.

Let's implement the flyweight design pattern:

1. Start by adding the Flyweight folder to the StructuralDesignPatters

folder. Now, add the Flyweight class:
public abstract class Flyweight {
public abstract void Operation(string extrinsicState);
}

2. This class is abstract and contains an abstract method called Operation(),

which is passed in the extrinsic state of the flyweight:
public class ConcreteFlyweight : Flyweight
{
public override void Operation(string extrinsicState)
{
Console.WriteLine($"ConcreteFlyweight: {extrinsicState}");
}
}

3. The ConcreteFlyweight class inherits the Flyweight class and overrides the
Operation() method. The method outputs the method name and its extrinsic
state. Now, add the FlyweightFactory class:
public class FlyweightFactory {
private readonly Hashtable _flyweights = new Hashtable();

public FlyweightFactory()
{
_flyweights.Add("FlyweightOne", new ConcreteFlyweight());
_flyweights.Add("FlyweightTwo", new ConcreteFlyweight());
_flyweights.Add("FlyweightThree", new ConcreteFlyweight());
}

public Flyweight GetFlyweight(string key) {

return ((Flyweight)_flyweights[key]);
}
}

[ 443 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

4. In our particular flyweight example, we store our flyweight objects in a hashtable.

Three flyweight objects are created in our constructor. Our GetFlyweight()
method returns the flyweight for the specified key from the hashtable. Now, add
the client:
public class Client
{
private const string ExtrinsicState = "Arbitary state can be
anything you require!";

private readonly FlyweightFactory _flyweightFactory = new

FlyweightFactory();

public void ProcessFlyweights()

{
var flyweightOne =
_flyweightFactory.GetFlyweight("FlyweightOne");
flyweightOne.Operation(ExtrinsicState);

var flyweightTwo =
_flyweightFactory.GetFlyweight("FlyweightTwo");
flyweightTwo.Operation(ExtrinsicState);

var flyweightThree =
_flyweightFactory.GetFlyweight("FlyweightThree");
flyweightThree.Operation(ExtrinsicState);
}
}

5. An extrinsic state can be anything you require it to be. In our example, we are
using a string. We declare a new flyweight factory, add three flyweights, and
execute the operation on each of them. Let's add the code to test our
implementation of the flyweight design pattern:
var flyweightClient = new
StructuralDesignPatterns.Flyweight.Client();
flyweightClient.ProcessFlyweights();

6. The code creates a new Client instance, and then calls the
ProcessFlyweights() method. You should see the following:

[ 444 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

Well, that's it for the structural patterns. Now it is time for us to look at implementing
behavioral design patterns.

Overview of behavioral design patterns

As a programmer, your behavior on the team is governed by your methods of
communication and interaction with other team members. The objects we program are no
different. As programmers, we determine how objects will behave and communicate with
other objects through the use of behavioral patterns. These behavioral patterns are as
follows:

Chain of responsibility: A sequential pipeline of objects that process an

incoming request.
Command: Encapsulates all the information that will be used to call a method at
some point in time within an object.
Interpreter: Provides interpretation of a given grammar.
Iterator: Use this pattern to access an aggregate object's elements sequentially
without exposing its underlying representation.
Mediator: Use this pattern to have objects communicate with each other via an
intermediary.
Memento: Use this pattern to capture and save the object's state.
Observer: Use this pattern to observe and be notified of changes in the object
state of the object being observed.
State: Use this pattern to alter the behavior of an object when its state changes.
Strategy: Use this pattern to define a catalog of encapsulated algorithms that are
interchangeable.
Template method: Use this pattern to define an algorithm and the steps that can
be overridden in subclasses.
Visitor: Use this pattern to add new operations to existing objects without
modifying them.

Due to the constraints of this book, we don't have enough pages left to cover the behavioral
design patterns. With that in mind, I will direct you to the following books, which you can
use to further your knowledge of design patterns. The first book is called Design Patterns in
C#: A Hands-on Guide with Real-World Examples, by Vaskaring Sarcar, and published by
Apress. The second book is called Design Patterns in .NET: Reusable Approaches in C# and F#
for Object-Oriented Software Design, by Dmitri Nesteruk, also published by Apress.
Published by Packt, the third book is called Hands-On Design Patterns with C# and .NET
Core, by Gaurav Aroraa and Jeffrey Chilberto.

[ 445 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

Between these books, you will not only come to understand all the patterns, but you will
also gain exposure to real-world examples, which will help you transition from simply
having head knowledge to having the practical skills to use design patterns in a reusable
way in your own projects.

That's it for our look at design pattern implementations. Before we summarize what we've
learned, I'll leave you with some final thoughts on clean code and refactoring.

Final thoughts
There are two types of software development—brownfield development and greenfield
development. The majority of the code we work on throughout our careers will be
brownfield development, which is the maintenance and extension of existing software,
while greenfield development is the development, maintenance, and extension of new
software. With greenfield software development, you are afforded the opportunity to write
clean code from the start, and I encourage you to do just that.

Make sure that projects are properly planned before you work on them. Then, employ the
tools available to you to develop clean code with confidence. When it comes to brownfield
development, you are best off spending time getting to know the system inside out before
you maintain or extend it. Unfortunately, you may not always be in a situation where time
affords you such luxury. So, there may be times when you will set about writing the code
you need, not realizing that code already exists to do the task you are implementing.
Keeping the code that you do write clean and well-structured will make for easier
refactoring later on in the project.

Regardless of whether the project you are working on is a brownfield or greenfield project,
it is down to you to ensure that you follow the company procedures. They are there for
good reasons, those reasons being harmony between the development team and a clean
code base. When you encounter unclean code within the code base, you should look to
refactor it immediately.

If the code is too complex to change immediately, and if too many changes across layers are
necessary, then the change must be logged as technical debt on the project to be addressed
at a later date after proper planning.

[ 446 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

At the end of the day, whether you call yourself a software architect, software engineer,
software developer, or anything else, for that matter, your bread and butter is your
programming skills. Bad programming can be detrimental to your current position, and can
even negatively impact your ability to find new positions. So, employ every resource you
have to ensure that your current code leaves a lasting good impression of your level of
ability. I once heard someone say the following:

"You are only as good as your last programming assignment!"

It is important when architecting systems not to be too clever and build overly complex
systems. Keep the depth of inheritance of your programs to no greater than 1, and do your
best to reduce loops through utilizing functional programming techniques such as LINQ.

You saw in Chapter 13, Refactoring C# Code – Identifying Code Smells, how LINQ is more
performant than a foreach loop. Try to also reduce the complexity of your software by
limiting the number of pathways through your computer program from the beginning to
the end. Reduce boilerplate code by removing the boilerplate code to aspects that can be
weaved into the code at compile time. This reduces the number of lines in your methods to
only those lines that are the required business logic. Keep classes small and focused on only
one responsibility. Also, keep methods to 10 lines of code or fewer. Classes and methods
must only perform a single responsibility.

Learn to keep the code you write simple so that it is easy to read and reason about.
Understand the code you write. If you can easily understand your code, then you're fine.
Now, ask yourself this: after working on another project and coming back to this one, would you
still understand the code with little or no effort? When code is hard to understand, then it must
be refactored and simplified.

Failure to do this can result in a bloated system that dies a slow and agonizing death. Use
documentation comments to document publicly accessible code. For hidden code, only use
succinct and meaningful comments when the code does not adequately make sense by
itself. Use patterns for common code that would often be repeated so that you Don't Repeat
Yourself (DRY). Indentation within Visual Studio 2019 is automatic, but the default
indentation is not the same across different document types. Therefore, it is a good idea to
make sure all documentation types have the same levels of indentation. Use the standard
naming recommendations as suggested by Microsoft.

Give yourself programming challenges to solve without copying and pasting other people's
source code. Use benchmarking (profiling) to rewrite the same code with the aim of
reducing processing time. Test your code often to ensure it is behaving and doing what it is
supposed to. Finally, practice, practice, and then practice some more.

[ 447 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

We all change our programming styles over time. Some programmers' code will deteriorate
over time if they are within a team of programmers that adopts a lot of poor practices.
Other programmers' code will improve over time if they are within a team of programmers
that adopts a lot of best practices. Don't forget, just because code compiles and does what it
is meant to, it does not necessarily mean that it is the cleanest or most performant code.

Your aim as a computer programmer is to write clean and efficient code that is easy to read,
reason, maintain, and expand. Practice implementing TDD and BDD, along with the
software paradigms of KISS, SOLID, YAGNI, and DRY.

Consider checking out some old code from GitHub to use as a training opportunity in
migration of old .NET versions to new .NET versions, and refactoring the code to make it
clean and performant, as well as adding documentation comments to produce API
documentation for the development team. This is good practice for honing your personal
computer programming skills. By doing this, you can often come across some rather clever
code that you can personally learn from. Other times, it can be a case of wondering what
the programmer was thinking at the time! But either way, improving your clean coding
skills at every opportunity you have will only work toward making you a stronger and
better programmer.

Another saying that I believe to be true in the field of programming is as follows:

"To become a true expert computer programmer, you have to push yourself beyond what
you are currently capable of doing."

So, no matter how expert you or your peers consider you to be, always remember that you
can do even better. Therefore, keep pushing forward and upping your game. Then, when
you retire, you can look back on your career with a righteous pride in your wonderful
accomplishments as a computer programmer!

Let's now summarise what we have learned in this chapter.

Summary
In this chapter, we covered several creational, structural, and behavioral design patterns.
You used the knowledge that you gained in this chapter to look at legacy code and
understand its goal. Then, you used the patterns that you learned to implement in this
chapter to refactor existing code and make it easier to read, reason, maintain, and extend.
By using the patterns in this book, and the many others that are available to you, you can
refactor existing code and write clean code from the start.

[ 448 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

You also used the creational design patterns to solve real-world problems and to improve
the efficiency of your code. Use structural design patterns to improve the overall structure
of code and improve relations between objects. Also, use behavioral design patterns to
improve communication between objects whilst maintaining the decoupling of those
objects.

Well, this is the end of the chapter, and I thank you for taking the time to read this book
and work through the code examples. Remember, software should be a joy to work with.
As such, we don't need unclean code causing problems for our business, its development
and support teams, and for the customers of the software. So, think about the code you are
writing, and always strive to be a better programmer than you are today—no matter how
many years you have been in the industry. There is an old saying: no matter how good you
are, you can always do better!

Let's test your knowledge on the contents of this chapter, and then I will leave you with
some further reading. Happy clean coding in C#!

Questions
1. What are GoF patterns and why would we use them?
2. Explain what creational design patterns are used for and list them.
3. Explain what structural design patterns are used for and them.
4. Explain what behavioral design patterns are used for and list them.
5. Is it possible to overuse design patterns and call code smells?
6. Describe the singleton design pattern and when you'd use it.
7. Why would we use factory methods?
8. What design pattern would you use to hide the complexity of a system that is
large and difficult to use?
9. How can you minimize memory usage and share common data between objects?
10. What pattern is used to decouple an abstraction from its implementation?
11. How can you construct multiple representations of the same complex object?
12. If you have an item that requires various stages of manipulation to get it into the
required state, what pattern would you use and why?

[ 449 ]
Refactoring C# Code – Implementing Design Patterns Chapter 14

Further reading
Refactoring: Improving the Design of Existing Code, by Martin Fowler
Refactoring at Scale, by Maude Lemaire
Software Development, Design, and Coding: With Patterns, Debugging, Unit Testing,
and Refactoring, by John F. Dooley
Refactoring for Software Design Smells, by Girish Suryanarayana, Ganesh
Samarthyam, and Tushar Sharma
Refactoring Databases: Evolutionary Database Design, by Scott W. Ambler and
Pramod J. Sadalage
Refactoring to Patterns, by Joshua Kerievsky
C#7 and .NET Core 2.0 High Performance, by Ovais Mehboob Ahmed Khan
Improving Your C# Skills, by Ovais Mehboob Ahmed Khan, John Callaway,
Clayton Hunt, and Rod Stephens
Patterns of Enterprise Application Architecture, by Martin Fowler
Working Effectively with Legacy Code, by Michael C. Feathers
https://www.dofactory.com/products/dofactory-net: C# Design Pattern
Framework for RAD by dofactory
Hands-On Design Patterns with C# and .NET Core, by Gaurav Aroraa and Jeffrey
Chilberto
Design Patterns Using C# and .NET Core, by Dimitris Loukas
Design Patterns in C#: A Hands-on Guide with Real-World Examples, by Vaskaring
Sarcar

[ 450 ]
Assessments

Chapter 1
1. One outcome of bad code is that you can end up with a really badly written piece
of code that is hard to understand. This can often lead to programmer stress and
software that is buggy, hard to maintain, and hard to test and extend.
2. One outcome of good code is that it is easy to read and understand, as you know
the programmer's intent. This leads to less stress for programmers who must
debug the code, test it, and extend it.
3. When you break a large project up into modular components and libraries, each
module can be worked on by separate teams concurrently. Small modules are
easy to test, code, document, deploy, extend, and maintain.
4. DRY stands for Don't Repeat Yourself. Look for repeatable code, and refactor it
so that you remove duplicate code. The advantage of this is smaller programs,
because if such code contains bugs, you only have to change it in one place.
5. KISS means simple code that will not confuse programmers, especially if you
have juniors on your team. KISS code is easy to read and write tests for.
6. S is the Single Responsibility Principle, O is the Open/Closed Principle, L is
Liskov Substitution, I is the Interface Segregation Principle, and D is
the Dependency Inversion Principle.
7. YAGNI is short for You Aren't Going to Need It. In other words, don't add code
you don't need. Only add the code you absolutely need, and no more.
8. Occam's Razor is the principle that states: Entities must not be multiplied without
necessity. Deal only in facts. Only make assumptions if absolutely necessary.

Chapter 2
1. The two roles in the peer code review are reviewer and reviewee.
2. The project manager agrees on the people that will be involved in the peer code
review.
Assessments

3. You can save your reviewer time and effort prior to requesting a peer code
review by making sure your code and tests all work, that you perform code
analysis on your project and fix any issues raised, and that your code adheres to
the company coding guidelines.
4. When reviewing code, look out for naming, formatting, programming styles,
potential bugs, correctness of code and tests, security, and performance issues.
5. The three categories of feedback are positive, optional, and critical.

Chapter 3
1. We can place our code in individual source files in folder structures and wrap
classes, interfaces, structs, and enums in namespaces that map to the folder
structure.
2. A class should have only one responsibility.
3. You can comment in your code for document generators using XML comments
placed directly above the public member to be documented.
4. Cohesion is the logical grouping together of code that works on the same
responsibility.
5. Coupling refers to the dependencies between classes.
6. Cohesion should be high.
7. Coupling should be low.
8. You can use DI and IoC to design for change.
9. DI stands for Dependency Injection.
10. IoC stands for Inversion of Control.
11. Immutable objects are type-safe and so can be safely passed between threads.
12. Objects should expose methods and properties and hide data.
13. Data structures should expose data and have no methods.

Chapter 4
1. Methods with no parameters are called niladic methods.
2. Methods with only one parameter are called monadic methods.
3. Methods with two parameters are called dyadic methods.
4. Methods with three parameters are called triadic methods.

[ 452 ]
Assessments

5. Methods with more than three parameters are called polyadic methods.
6. You should avoid duplicate code. It is not a productive way to program, can
make programs unnecessarily large, and has the propensity to proliferate the
same exception throughout your codebase.
7. Functional programming is a software coding methodology that treats
computations as the mathematical evaluation of computations that does not
modify state.
8. The advantages of functional programming include safe code in multithreaded
applications and smaller, more meaningful methods that are easy to read and
understand.
9. Input and output can be a problem for functional programs as it relies on side-
effects. Functional programming does not allow for side-effects.
10. WET code is the opposite of DRY in that code is written each time it is needed.
This produces duplication, and the same exception can occur in multiple
locations within a program, making maintenance and support more difficult.
11. DRY code is the opposite of WET in that code is only ever written once and is
reused wherever it is needed. This reduces the code base and exception footprint,
thus making programs easier to read and maintain.
12. You DRY out WET code by removing duplicate code using refactoring.
13. Long methods are cumbersome and prone to exceptions. The smaller they are,
the easier they are to read and maintain. There is also less chance of the
programmer introducing bugs, especially of a logical nature.
14. To avoid having to use try/catch blocks, you can write argument validators. You
would then call the validators at the top of your method. If the parameters fail
validation, then the appropriate exception is thrown, and the method is not
executed.

Chapter 5
1. A checked exception is an exception that is checked at compile time.
2. An unchecked exception is an exception that is not checked or simply ignored at
compile time.
3. An overflow exception is raised when high-order bits cannot be assigned to the
destination type. In checked mode, OverflowException is raised. In unchecked
mode, high-order bits that cannot be assigned are simply ignored.
4. An attempt made to access a property or a method on a null object.

[ 453 ]
Assessments

5. Implement a Validator class and an Attribute class that checks the parameter
for null, and that throws ArgumentNullException. You would use the
Validator class at the top of your methods so that you don't get halfway
through the method before the exception is raised.
6. Business Rule Exception (BRE).
7. BREs are bad practice because they expect exceptions to be raised in order to
control program flow.
8. Correct programming should never control the flow of a computer program by
expecting exceptions as output. So, given that BREs are bad as they expect
exceptional output and use it to control program flow, a better solution is to use
conditional programming. With a conditional program, you use Boolean logic.
Boolean logic allows for two possible paths of execution, and never raises
exceptions. Conditional checks are explicit and make the programs easier to read
and maintain. You can also easily extend such code, whereas with BREs, you
can't.
9. First, start with error trapping for known types of exceptions such as
ArgumentNullExceptions and OverflowExceptions using known exception
types in the Microsoft .NET Framework. But when these are insufficient and
don't provide enough data for your particular situation, then you would write
and use your own custom exceptions and apply meaningful exception messages.
10. Your custom exception must inherit from System.Exception, and implement
three constructors: the default construct, a constructor that accepts a text
message, and a constructor that accepts a text message and an inner exception.

Chapter 6
1. A good unit test must be atomic, deterministic, repeatable, and fast.
2. A good unit test must not be inconclusive.
3. Test-driven development.
4. Behavioral-driven development.
5. A small unit of code whose only purpose is to test a single unit of code that only
does one thing.
6. A fake object used by the unit test to test the public methods and properties of a
real object, but without testing the method or property dependencies.
7. A fake object is the same as a mock object.
8. MSTest, NUnit, and xUnit.

[ 454 ]
Assessments

9. Rhino Mocks and Moq.

10. SpecFlow.
11. Unnecessary comments, dead code, and redundant tests.

Chapter 7
1. The testing of a complete system from end to end. This can be performed
manually, automatically, or by using both methods.
2. Integration testing.
3. Manual testing of all features, all our unit tests should pass, and we should write
automation tests to test the commands and data that are passed between two
modules.
4. Factories are classes that implement the factory method pattern whose intention
is to allow the creation of objects without specifying their classes. We would use
them in the following scenarios:
1. The class is unable to anticipate the type of object that must be
instantiated.
2. The subclass must specify the type of object to instantiate.
3. The class controls the instantiation of its objects.
5. DI is a method of producing loosely coupled code that is easy to maintain and
extend.
6. Using a container makes the management of dependency objects easy.

Chapter 8
1. A thread is a process.
2. One.
3. Background threads and foreground threads.
4. The background thread.
5. The foreground thread.
6. Thread.Sleep(500);
7. var thread = new Thread(Method1);

[ 455 ]
Assessments

8. Set IsBackground equal to true.

9. A deadlock is a situation when two threads are blocked and waiting on the other
thread to release the resource.
10. Monitor.Exit(objectName);
11. Multiple threads using the same resource generate different outputs based on the
timings of each thread.
12. Use the TPL with ContinueWith(), and use Wait() to wait until the task has
finished before exiting the method.
13. Using a member variable that is shared by other methods, and passing in
reference variables.
14. Yes.
15. The ThreadPool.
16. It is an object that cannot be modified once it has been constructed.
17. They allow you to safely share data between threads.

Chapter 9
1. Application Programming Interface.
2. Representational State Transfer.
3. Uniform interface, client-server, stateless, cacheable, layered system, optional
executable code.
4. Hypermedia as the Engine of Application State (HATEOAS).
5. RapidApi.com.
6. Authorization and authentication.
7. Claims are statements that an entity makes about itself. These claims are then
validated against a data store. They are particularly useful in role-based security
to check whether the entity making the claim is authorized in regard to that
claim.
8. Making API requests and examining their responses.
9. Because you can change your data store in keeping with your requirements.

[ 456 ]
Assessments

Chapter 10
1. The correct partitioning of software into logical namespaces, interfaces, and
classes, which aids the testing of software.
2. By understanding APIs, you can KISS your code and keep it DRY by not
reinventing the wheel and writing code that already exists. This saves time,
energy, and money.
3. Structs.
4. Third-party APIs are written by software developers, and so subject to human
error that introduces bugs. By testing third-party APIs, you can be confident they
work as expected, and if not, then you can have the code fixed or write a wrapper
for it.
5. Your APIs are prone to errors. By testing them in keeping with the specification
and its acceptance criteria, you can be sure you are delivering what the business
wants at the agreed level of quality ready for public release.
6. The specification and acceptance criteria provide the normal program flow. From
them, you can determine what to test in regard to the normal flow of execution,
and you can determine what exceptional circumstances will be encountered and
test for them.
7. Namespaces, interfaces, and classes.

Chapter 11
1. Cross-cutting concerns are those concerns that not part of the business
requirements that form the core concerns, but that must be addressed in all areas
of the code. AOP stands for Aspect-Oriented Programming.
2. An aspect is an attribute that, when applied to a class, method, property, or
parameter, injects code at compile time. You apply an aspect in square brackets
before the item it is being applied to.
3. An attribute gives semantic meaning to an item. You apply an attribute in square
brackets before the item it is being applied to.
4. Attributes give the code semantic meaning, while aspects remove the boilerplate
code so that it is injected at compile time.
5. When the code is being built, the compiler will insert the boilerplate code that the
aspect hides from the programmer. This process is known as code weaving.

[ 457 ]
Assessments

Chapter 12
1. Code metrics are several source code measurements that enable us to identify
how complex our software is, and how maintainable it is. Such measurements
enable us to identify areas of code that can be made less complex and more
maintainable through refactoring.
2. Cyclomatic complexity, maintainability index, depth of inheritance, class
coupling, lines of source code, and lines of executable code.
3. Code analysis is the static analysis of source code with the intention of
identifying design flaws, issues with globalization, security problems, issues with
performance, and interoperability problems.
4. Quick actions are single commands identified by a screwdriver or lightbulb that
will suppress warnings, add using statements, import missing libraries and add
the using statements, correct errors, and implement language usage
improvements aimed at simplifying code and reducing the number of lines in a
method.
5. JetBrains' dotTrace utility is a profiling tool used for the purpose of profiling
source code and compiled assemblies to identify potential issues with the
software. With it you can perform sampling, tracing, line-by-line, and timeline
profiling. You can profile execution time, thread time, real-time CPU
instructions, and thread cycle time.
6. JetBrains' ReSharper utility is a code refactoring tool that helps developers
identify and fix code issues and implement language features to improve and
speed up the programmer's programming experience.
7. The decompilation of source code can be used to retrieve lost source code,
generate PDBs for debugging, and for learning. You can also use the decompiler
to see how well you have obfuscated your code to make it hard for hackers and
other people to steal your code secrets.

Chapter 13
1. Application-level, class-level, and method-level.
2. Boolean blindness, combinatorial explosion, contrived complexity, data clump,
deodorant comments, duplicate code, lost intent, mutation of variables, oddball
solution, shotgun surgery, solution sprawl, and uncontrolled side effects.

[ 458 ]
Assessments

3. Cyclomatic complexity, divergent change, downcasting, excessive literal use,

feature envy, inappropriate intimacy, indecent exposure, large class (also known
as God object), lazy class (also known as freeloader and lazy object), middleman
class, an orphan class of variables and constants, primitive obsession, refused
bequest, speculative generality, Tell, don't ask!, and temporary field.
4. Black sheep, cyclomatic complexity, contrived complexity, dead code, excessive
data return, feature envy, identifier size, inappropriate intimacy, long line aka
God line, lazy method, long method (God method), long parameter list (too
many parameters), message chains, middleman method, oddball solutions, and
speculative generality.
5. Use LINQ instead of loops. Make classes responsible for only one thing. Make
methods do only one thing. Replace long lists of parameters with parameter
objects. Use creational design patterns to improve the efficiency of expensive
object creation and utilization. Keep methods to 10 lines or less. Use AOP to
remove boilerplate code from methods. Decouple objects and make them
testable. Make code highly cohesive.
6. A value that represents the amount of branching and looping.
7. Reduce the amount of branching and looping that takes place until such time as
the cyclomatic complexity value becomes 10 or less.
8. Making things more complicated than they need to be.
9. Keep It Simple, Stupid (KISS).
10. The same thing is done by different methods with different parameter
combinations.
11. Create generic methods that can perform the same task on the different data
types so that you only have the one method with one set of parameters.
12. Fix the bad code and remove the comment.
13. Ask for help.
14. Stack Overflow.
15. A long parameter list can be replaced with a parameter object.
16. Refactor it into smaller methods that do only one thing, and remove boilerplate
code into aspects using AOP.
17. No more than 10 lines.
18. 0-10; anything beyond that and you are asking for trouble.
19. One.
20. Variables, classes, properties, and methods that are not used. Get rid of them.
21. Choose the best method of implementation, and then refactor the code to use just
that method of implementation.

[ 459 ]
Assessments

22. Refactor the temporary field and the methods that operate on it into their own
class.
23. The same set of variables used in different classes. Refactor the variables into a
class of their own, and then reference the class.
24. A class inherits from another class but does not use all its methods.
25. The Law of Demeter.
26. Only allow classes to speak to their immediate neighbors.
27. A class or method spending too much time inside another class or method.
28. Refactor dependencies in their own class or method.
29. The factory method.
30. Inherit from a base class, and then create the new class that inherits from the
base.
31. Single responsibility is implemented in different methods of different classes
across different layers of the application. Refactor the responsibility into its own
class so that it is only in a single location.
32. The data should be placed in the same object that operates on it.
33. When you create an object that asks another object for data so that it can perform
operations on it.
34. A single change requires changes in multiple locations. Remove duplication,
remove the coupling, and improve cohesion.
35. Lost intent is when the reason for the class or method is unclear because there are
lots of unrelated items clumped together. Refactor the code so that all methods
are in the right class. That way, the intent of the class and the methods becomes
clear.
36. You can refactor loops with LINQ queries. LINQ is a functional language that
does not alter location variables and can perform much faster than loops.

Chapter 14
1. GoF is short for Gang-of-Four patterns. These are 23 patterns that are grouped
into creational, structural, and behavioral design patterns. They are considered
the foundation of all software design patterns. They work together to produce
clean object-oriented code.
2. Creational patterns enable abstraction and inheritance to provide an object-
oriented way of removing code duplication and improving performance when
object creation is expensive. The creational patterns are abstract factory, factory
method, singleton, prototype, and builder.

[ 460 ]
Assessments

3. Structural patterns enable the correct management of relationships between

objects. We can use structural patterns to enable incompatible interfaces to work
together, decouple abstractions from their implementations, and improve
performance. The structural patterns are adapter, bridge, composite, decorator,
façade, flyweight, and proxy.
4. Behavioral patterns govern how objects interact and communicate with each
other. We can use them to produce pipelines, encapsulate commands and
information to be executed at a future point in time, mediate between objects,
observe state changes in objects, and more. The behavior patterns are chain of
responsibility, command, interpreter, iterator, mediator, memento, observer,
state, strategy, template method, and visitor.
5. Yes.
6. The singleton only allows a single instance of an object throughout the lifetime of
the application. The object is globally accessible to all objects that need it. We use
this pattern when we need to ensure we have one centralized point of object
creation and object access.
7. We use factory methods when we have a need to create objects without
specifying the exact class to be instantiated.
8. Façade.
9. Use the flyweight design pattern.
10. Bridge.
11. Use the builder pattern.
12. You would use the chain of responsibility pattern, as you can have a pipeline of
handlers, each of which performs a task. If they are unable to handle the task, the
handlers pass the task to their successor to handle.

[ 461 ]
Other Books You May Enjoy
If you enjoyed this book, you may be interested in these other books by Packt:

Hands-On Software Architecture with C# 8 and .NET Core 3

Francesco Abbruzzese, Gabriel Baptista
ISBN: 978-1-78980-093-7
Overcome real-world architectural challenges and solve design consideration
issues
Apply architectural approaches like Layered Architecture, service-oriented
architecture (SOA), and microservices
Learn to use tools like containers, Docker, and Kubernetes to manage
microservices
Get up to speed with Azure Cosmos DB for delivering multi-continental
solutions
Learn how to program and maintain Azure Functions using C#
Understand when to use test-driven development (TDD) as an approach for
software development
Write automated functional test cases for your projects
Other Books You May Enjoy

Learn C# Programming
Marius Bancila, Raffaele Rialdi, Ankit Sharma
ISBN: 978-1-78980-586-4
Get to grips with all the new features of C# 8
Discover how to use attributes and reflection to build extendable applications
Utilize LINQ to uniformly query various sources of data
Use files and streams and serialize data to JSON and XML
Write asynchronous code with the async-await pattern
Employ .NET Core tools to create, compile, and publish your applications
Create unit tests with Visual Studio and the Microsoft unit testing frameworks

[ 463 ]
Other Books You May Enjoy

Leave a review - let other readers know what

you think
Please share your thoughts on this book with others by leaving a review on the site that you
bought it from. If you purchased the book from Amazon, please leave us an honest review
on this book's Amazon page. This is vital so that other potential readers can see and use
your unbiased opinion to make purchasing decisions, we can understand what our
customers think about our products, and our authors can see your feedback on the title that
they have worked with Packt to create. It will only take a few minutes of your time, but is
valuable to other potential customers, our authors, and Packt. Thank you!

[ 464 ]
Index

deodorant comment 391

A duplicate code 392
abstract factory pattern lost intent 392, 393
implementing 422, 423, 424, 425 mutation of variables 393, 394, 395
Adapter Pattern oddball solution 395, 396, 397
using 395 shotgun surgery 397, 398, 399
admin module (subsystem) 179, 180 solution sprawl 399
advice 334 uncontrolled side effects 399
agnostic RAML design specification arguments 100
C# API, generating from 277, 278, 280, 281 Aspect-Oriented Programming (AOP), with
API development PostSharp
Swagger for 262, 263, 264 architectural framework, extending 337, 338
API documentation aspect, developing 334
quality, need for 261, 262 behaviors, injecting before and after method
API endpoints 258 execution 334, 335, 336, 337
API proxy 254, 255, 256 framework, extending 334
API Workbench Aspect-Oriented Programming (AOP)
installing, by MuleSoft 272, 273 about 33, 327, 392
Application Lifecycle Management (ALM) 360 with PostSharp 333
Application Programming Interfaces (APIs), design Atom
guidelines installation link 272
API documentation, quality need for 261, 262 installing, by MuleSoft 272, 273
immutable structs, passing instead of mutable attribute programming 327
objects 265, 266, 267 Azure Active Directory (Azure AD) 288
well-defined software boundaries 259, 260, 261 Azure Key Vault
Application Programming Interfaces (APIs) Morningstar API key, storing 286, 287, 288, 289
about 251, 252, 253, 254, 374 Azure
design guidelines 256, 257, 258, 259 dividend calendar ASP.NET Core web
designing, with RAML 271 application, creating 289, 290
project, creating 274, 276 home page, reference link 296
testing 269, 270, 271 reference link 286
application-level code smells
about 387 B
Boolean blindness 387, 388, 389
background threads 206
combinatorial explosion 389, 390
bad code
contrived complexity 390, 391
versus good code 9
data clump 391
bad coding practices
about 9 specification 277, 278, 280, 281
bad naming conventions 13 C# exception handling
code, that is difficult to read 19, 20 best practices 127, 128
code, that is tightly coupled 20, 21 C# programming
commented-out lines 12 multiple parameters, removing 100, 101
comments, that excuse bad code 11 central processing unit (CPU) 358
comments, that state the obvious 10, 11 checked exceptions 110, 111, 112, 113, 114
directly expose information 24 class coupling 361
exceptions, using to control program flow 18, 19 class-level code smells, cyclomatic complexity
Finalize() method, using 22 readability, improving of conditional checks within
improper indentation 10 if statement 403, 404
improper organization, of namespaces 12 switch statements, replacing with factory pattern
Keep It Simple, Stupid (KISS) 22, 23 400, 401, 402, 403
large classes, lack of regions 23 class-level code smells
lost-intention code 23, 24 about 400
low cohesion 21 cyclomatic complexity 400
methods 16, 17 divergent change 404, 405
methods, with more than 10 lines of code 17 downcasting 405
methods, with more than two parameters 18 feature envy 405, 407
multiple jobs, classes 13, 14, 15, 16 freeloading class 408, 409
objects, left hanging around 21, 22 God object 408
over-engineering 22 inappropriate intimacy 407
behavioral design patterns indecent exposure 408
overview 445, 446 literal, using 405
Behavioral-Driven Development (BDD) 33, 131 middleman class 409
binary orphan class, of variables and constants 409
reference link 218 primitive obsession 409
Boolean blindness 387, 388, 389 refused bequest 410
bounded context 259 speculative generality 410
bridge pattern Tell, Don't Ask 410
implementing 434, 435, 436 temporary fields 410
brownfield development 446 classes
builder pattern organizing 62, 64, 65
implementing 427, 428, 429, 430, 431, 432, responsibility 65, 66, 67
433 code analysis
business requirement performing 363, 364, 365, 366
designing 75 code block 98
Business Rule Exceptions (BREs) code cleanup
about 18, 117, 118, 119, 120 performing 360, 361, 362, 363
used, for handling conditions within program code metrics
120, 121 calculating 360, 361, 362, 363
code review feedback
C effects 48, 49, 50
C# API providing 57
generating, from agnostic RAML design providing, as reviewer 57, 58

[ 466 ]
responding to 57 coupling
responding to, as reviewee 58 low coupling, example 72, 73
code review process 39, 40 tight coupling, example 71
code review, aspects creational design patterns
about 51, 55 abstract factory pattern 422, 423, 424, 425
architectural guidelines 54 builder pattern 427, 428, 429, 430, 431, 432,
business requirements 51 433
company's coding guidelines 51 factory method pattern 420, 421, 422
design patterns 54 implementing 418
formatting 52 prototype pattern 425, 426, 427
naming conventions 51 singleton design pattern 419, 420
performance 54 critical feedback 58
security 54 cross-cutting concerns reusable library
testing 52, 53 about 338
code review caching concern, adding 338, 339
leading 41, 42 configuration settings concern, adding 354, 355
performing 55, 56 exception-handling concern, adding 343, 344
preparing for 40, 41 instrumentation concern, adding 355, 356
pull request, issuing 43, 44, 45 logging capabilities, adding 340, 341
pull request, responding to 46, 47, 48 logging concern, adding 341, 342
code smell 386 resource pool concern, adding 353, 354
coding conventions security concern, adding 344, 345, 346, 347
about 33 transaction concern, adding 353
reference link 33 validation concern, adding 348, 349, 350, 352
coding methodologies custom exceptions
need for 31 building 125, 126, 127
coding principles cyclomatic complexity 360
about 32
need for 31 D
coding standards data clump 391
about 31, 32 data structure
need for 31 data, exposing 88
references 32 example 89
cohesion 70 data transfer objects (DTOs) 93
high cohesion, example 74, 75 data
low cohesion, example 73 exposing, in data structure 88
combinatorial explosion 389, 390 hiding, in objects 87
comments dead code
removing 169, 170 about 169
Common Language Runtime (CLR) 21 removing 169, 170
composite pattern deadlock example
implementing 436, 437, 438, 439 coding 220, 221, 222, 223, 224, 225
concurrency 204 deadlocks
contrived complexity 390, 391 about 219
core concerns 326 preventing 219

[ 467 ]
decorator design pattern 327, 329, 330 subsystems 175
deodorant comment 391 encapsulation
Dependency Injection (DI) example 87, 88
about 79, 194, 195, 196, 197, 199, 200 End-to-end (E2E)
example 79, 80, 81 about 173
types 194 used, for testing three-module system 182, 183,
Depth of Inheritance 361 184, 185
divergent change 404, 405 exceptions 123, 124
dividend calendar API, securing with Morningstar
API key F
authentication, adding 300, 301, 302, 303, 304 factories
authentication, setting up 300 implementing, with factory method pattern 185,
authorization, adding 304, 305, 306, 307 186, 187, 188, 189, 190, 191, 192, 193, 194
authorization, setting up 300 factory method pattern
repository, setting up 297, 298, 299 implementing 420, 421, 422
dividend calendar API, securing used, for implementing factories 185, 186, 187,
Morningstar API key, using 297 188, 189, 190, 191, 192, 193, 194
dividend calendar API façade pattern
about 284, 285 implementing 439, 440, 441
throttling 319, 320, 321, 322, 323 feature envy 405
dividend calendar ASP.NET Core web application flyweight pattern
creating, in Azure 289, 290 implementing 442, 443, 444, 445
publishing 291, 292, 293, 294, 295, 296, 297 foreground threads 206
dividend calendar code functional programming
adding 311, 312, 313, 314, 315, 316, 317, about 92, 93, 94, 95
318, 319 code, indenting 97, 98, 99
document generation keeping, methods small 95, 96, 97
commenting 67, 69, 70
Domain-Driven Design (DDD) 259 G
domains 259
Gang-of-Four (GoF) 54, 417
Don't Repeat Yourself (DRY) 32, 35, 99, 261, 447
Garbage Collection (GC) 21
downcasting 405
God method 102
DRY code
God object 102
duplication, removing 99
good code
duplicate code 392
versus bad code 9
dyadic methods 100
good coding practices
about 24
E API documentation comments 25
E2E testing, subsystems classes, that only do one job 27
admin module (subsystem) 179, 180 code, that is loosely coupled 29
login module (subsystem) 175, 176, 177, 178, code, that is readable 28
179 Dispose() method, using 29, 30
test module (subsystem) 181, 182 exceptions, using properly 28
E2E testing Finalize() method, avoiding 30
about 173, 174, 175 good naming conventions 26, 27

[ 468 ]
high cohesion 29 Inversion of Control (IoC)
meaningful comments 25 about 79
methods, that do one thing 27 example 81, 82
methods, with 4 lines 27
methods, with less than 10 lines 27 J
methods, with no more than two parameters 28 JavaScript Object Notation (JSON) 285
namespaces, using for proper organization 26 JetBrains dotTrace profiler
proper indentation 25 using 367, 368, 369, 370, 371, 372
regions, using in large classes 31 JetBrains ReSharper
right level of abstraction 30 using 372, 373, 374, 375, 376, 377, 378, 379,
good-quality code 358, 359 380, 381, 382
greenfield development 446
K
H Keep It Simple, Stupid (KISS) 22, 23, 32, 34, 261,
Her Majesty's Revenue and Customs (HMRC) 219 390
high cohesion
about 70 L
example 74, 75 Law of Demeter
HTTP status codes about 83
URL 257 example 83, 84, 85
Human Resources Index (ihridx) 13 login module (subsystem) 175, 176, 177, 178,
Hypermedia as the Engine of Application State 179
(HATEOAS) 253 London Stock Exchange (LSE) 127
HyperText Transfer Protocol (HTTP) 367 low cohesion
about 70
I example 73
immutability low coupling
demonstrating 236, 238 example 72, 73
immutable data structure type
about 85 M
example 86 Maintainability Index 360
immutable object type Market Identification Code (MIC) 313
about 85 method-level code smells
example 86 about 411
immutable types 234 black sheep method 411
indecent exposure 408 contrived complexity 411
input/output (I/O) 358 cyclomatic complexity 411
integration testing 173, 202 dead code 411
Interface-Oriented Programming (IOP) 76, 77, 78, excessive data return 412
79 feature envy 412
Interlocked class God lines 413
using 243, 244, 245, 246, 247 God methods 413
Intermediate Language (IL) 359 identifier size 412
International Resource Identifier (IRI) 256 inappropriate intimacy 412
Internet Information Services (IIS) 368 lazy method 413

[ 469 ]
long parameter lists 413
message chain 413 O
middleman method 414 Object-Oriented Programming (OOP) 19, 361
oddball solution 414 objects
speculative generality 414 data, hiding 87
methods methods, exposing 87
exposing, in objects 87 Occam's razor 36
Model-View-ViewModel (MVVM) 333 oddball solution 395, 396, 397
modularity 33 optional feedback 58
modularization 200, 201, 202
Monadic methods 100 P
Moq 152, 153, 154, 155, 156, 157 Peer Review (PR) 57
Morningstar API key Performance, Availability, Security, Scalability,
security, testing 308, 309, 310, 311 Maintainability, Accessibility, Deployability, and
storing, in Azure Key Vault 286, 287, 288, 289 Extensibility (PASSMADE) 358
used, for securing dividend calendar API 297 polyadic methods 100
Morningstar API positive feedback 58
accessing 286 PostSharp
reference link 286 Aspect-Oriented Programming (AOP) with 333
MSTest Framework URL 333
installing 137, 139, 140, 141, 142, 143, 144, principle of least astonishment (POLA) 85
145 process 204
MuleSoft prototype pattern
Atom and API Workbench, installing by 272, 273 implementing 425, 426, 427
multi-threaded applications proxy pattern 331, 332, 333
general recommendations, from Microsoft 247, pull request
248 issuing 43, 44, 45
multi-threading 204 responding to 46, 47, 48
mutability
about 234 Q
demonstrating 234, 236
Quick Action tool
Mutual Exclusion Objects (mutexes)
using 366, 367
about 212
benefits 213
disadvantages 213
R
using, with synchronous threads 212, 214, 215 race conditions
preventing 225, 226, 227
N RAML
Application Programming Interfaces (APIs),
Niladic methods 100
designing with 271
normal program flow
URL 271, 281
used, for handling conditions within program
redundant tests
121, 122, 123
removing 169, 170
NullPointerExceptions
refused bequest 410
avoiding 114, 115, 116, 117
repository pattern 297
NUnit 145, 146, 148, 150, 151, 152
Representational State Transfer (REST) 252

[ 470 ]
resource 258 synchronous threads
REST services mutex, using with 212, 214, 215
constraints 253
RESTful APIs 252 T
Task Parallel Library (TPL)
S about 208, 209
semaphores Parallel.For() 210, 211
about 215 Parallel.Invoke() 209, 210
advantages 215 Telerik JustDecompile
used, for working with parallel threads 215, 216, using 382, 383, 384
217 test module (subsystem) 181, 182
shotgun surgery 397, 398, 399 Test-Driven Development (TDD)
Single Responsibility Principle (SRP) about 33, 64, 131
about 14, 65, 92 methodology practice 162, 163, 165, 166, 168,
implementing 103, 104, 105, 106, 107 169
Single Responsibility Principle (SRP)SRP) testing tools
implementing 102 about 137
Single Responsibility Principle, Open-Closed Moq 152, 153, 154, 155, 156, 157
Principle, Liskov Substitution, Interface MSTest Framework 137, 139, 140, 141, 142,
Segregation Principle, and Dependency 143, 144, 145
Inversion Principle (SOLID) NUnit 145, 146, 148, 150, 151, 152
about 32, 35, 36, 261 SpecFlow 157, 159, 160, 162
principles 35 third-party APIs
Single-Page Applications (SPAs) 259 testing 268, 269
singleton design pattern thread parameters
implementing 419, 420 adding 207, 208
Software Development Life Cycle (SDLC) 56 thread pool
SpecFlow number of processors and threads, limiting 217,
about 157, 159, 160, 162 218
URL 162 using 208
ssynchronized method dependencies 243 thread safety 234, 238, 239, 241, 242
static constructors ThreadPool.QueueUserWorkItem() method 212
about 228 threads
adding, to sample code 229, 230 ackground threads 206
static methods foreground threads 206
about 228 life cycle 205, 206, 207
adding, to sample code 230, 231, 232, 233 three-module system
structural design patterns testing, with End-to-end (E2E) 182, 183, 184,
bridge pattern 434, 435, 436 185
composite pattern 436, 437, 438, 439 tight coupling
façade pattern 439, 440, 441 about 70
flyweight pattern 442, 443, 444, 445 example 71
implementing 433 time slicing 204
Structured Query Language (SQL) 367 time the software is functional (tsf) 358
synchronized method dependencies 242 total time it is expected to function (ttef) 358

[ 471 ]
triadic methods 100
type initializer 228 W
WET code
U duplication, removing 99, 100
Windows Communication Foundation (WCF) 368
unbounded priority inversion 215
wrapper class 269
unchecked exceptions 110, 111, 112, 113, 114
Write Every Time (WET) 99
unit testing
need for 131, 132, 133, 135, 136, 137
Universal Windows Platform (UWP) 368 Y
unmanaged code (COM) 21 You Ain't Gonna Need It (YAGNI) 32, 34, 261
User Acceptance Testing (UAT) 56

Hajian M. Flutter Engineering 2024
100% (1)
Hajian M. Flutter Engineering 2024
691 pages
Ebenezer D. Simplified JavaScript For Very Important Programmers 2023
No ratings yet
Ebenezer D. Simplified JavaScript For Very Important Programmers 2023
226 pages
Mellor A Java Oop Done Right Create Object Oriented Code You
No ratings yet
Mellor A Java Oop Done Right Create Object Oriented Code You
202 pages
Design Patterns by Tutorials (2nd Edition) - 2018
100% (1)
Design Patterns by Tutorials (2nd Edition) - 2018
364 pages
Backend Developer in 30 Days: Acquire Skills on API Designing, Data Management, Application Testing, Deployment, Security and Performance Optimization (English Edition)
From Everand
Backend Developer in 30 Days: Acquire Skills on API Designing, Data Management, Application Testing, Deployment, Security and Performance Optimization (English Edition)
Pedro Marquez-Soto
No ratings yet
C & C++ Interview Questions You'll Most Likely Be Asked
From Everand
C & C++ Interview Questions You'll Most Likely Be Asked
Vibrant Publishers
No ratings yet
Clean Code Javascript
No ratings yet
Clean Code Javascript
46 pages
Cracking C Programming Interview: 500+ interview questions and explanations to sharpen your C concepts for a lucrative programming career (English Edition)
From Everand
Cracking C Programming Interview: 500+ interview questions and explanations to sharpen your C concepts for a lucrative programming career (English Edition)
Tanuj Kumar Jhamb
No ratings yet
Question .Net Core
No ratings yet
Question .Net Core
8 pages
React Hooks Cheat Sheet: Usestate Usereducer
100% (4)
React Hooks Cheat Sheet: Usestate Usereducer
1 page
C# in Depth, 3rd Ed
0% (1)
C# in Depth, 3rd Ed
1 page
Professional C++
From Everand
Professional C++
Marc Gregoire
2/5 (3)
Implementing Design Patterns in C# and .NET 5: Build Scalable, Fast, and Reliable .NET Applications Using the Most Common Design Patterns (English Edition)
From Everand
Implementing Design Patterns in C# and .NET 5: Build Scalable, Fast, and Reliable .NET Applications Using the Most Common Design Patterns (English Edition)
Alexandre F. Malavasi Cardoso
No ratings yet
Socket.IO Cookbook: Over 40 recipes to help you create real-time JavaScript applications using the robust Socket.IO framework
From Everand
Socket.IO Cookbook: Over 40 recipes to help you create real-time JavaScript applications using the robust Socket.IO framework
Tyson Cadenhead
4/5 (1)
Road To Learn React
50% (2)
Road To Learn React
226 pages
Typescript Tutorial
100% (5)
Typescript Tutorial
189 pages
Java Design Patterns With Examples
96% (25)
Java Design Patterns With Examples
92 pages
High Performance Enterprise Apps using C# 10 and .NET 6: Hands-on Production-ready Clean Codes, Pattern Matching, Benchmarking, Responsive UI and Performance Tuning Tools (English Edition)
From Everand
High Performance Enterprise Apps using C# 10 and .NET 6: Hands-on Production-ready Clean Codes, Pattern Matching, Benchmarking, Responsive UI and Performance Tuning Tools (English Edition)
Ockert J. du Preez
No ratings yet
Angular Form Essentials
100% (1)
Angular Form Essentials
133 pages
Mastering TypeScript - Sample Chapter
No ratings yet
Mastering TypeScript - Sample Chapter
33 pages
MongoDB and Mongoose in NodeJS
100% (3)
MongoDB and Mongoose in NodeJS
35 pages
Alfresco 3 Cookbook
From Everand
Alfresco 3 Cookbook
Snig Bhaumik
No ratings yet
Node - Js Design Patterns Sample Chapter
100% (1)
Node - Js Design Patterns Sample Chapter
56 pages
Power of Simplicity A Quick Tour of The Dotnet Cli How To Test Asp. Net Core Web Api
No ratings yet
Power of Simplicity A Quick Tour of The Dotnet Cli How To Test Asp. Net Core Web Api
56 pages
Modern C++ Programming: Including the recent standards C++11, C++17, C++20, C++23
From Everand
Modern C++ Programming: Including the recent standards C++11, C++17, C++20, C++23
Orhan Gazi
No ratings yet
Mastering C# 8.0: Master C# Skills with Hands-on Code Examples (English Edition)
From Everand
Mastering C# 8.0: Master C# Skills with Hands-on Code Examples (English Edition)
Joydip Kanjilal
No ratings yet
Mastering Node - Js
No ratings yet
Mastering Node - Js
25 pages
Learning F# Functional Data Structures and Algorithms - Sample Chapter
No ratings yet
Learning F# Functional Data Structures and Algorithms - Sample Chapter
26 pages
C The Ultimate Beginners Guide To C Programming With Step by Step
100% (1)
C The Ultimate Beginners Guide To C Programming With Step by Step
87 pages
Jean Paul V.A - Design Patterns in C# - 2012
100% (2)
Jean Paul V.A - Design Patterns in C# - 2012
94 pages
C# Tutorial
No ratings yet
C# Tutorial
1 page
Advanced Javascript Unleashed
No ratings yet
Advanced Javascript Unleashed
234 pages
Learn C# Programmin
No ratings yet
Learn C# Programmin
10 pages
C#.Net Interview Questions and Answers
No ratings yet
C#.Net Interview Questions and Answers
11 pages
Practical C++ Backend Programming: Crafting Databases, APIs, and Web Servers for High-Performance Backend
From Everand
Practical C++ Backend Programming: Crafting Databases, APIs, and Web Servers for High-Performance Backend
Justin Barbara
No ratings yet
Backend Developer - Java
No ratings yet
Backend Developer - Java
2 pages
Dart
100% (2)
Dart
48 pages
Namaste Frontend System Design Curriculum
No ratings yet
Namaste Frontend System Design Curriculum
12 pages
Learning Dart
From Everand
Learning Dart
Ivo Balbaert
No ratings yet
Go (Programming Language) - Wikipedia, The Free Encyclopedia
No ratings yet
Go (Programming Language) - Wikipedia, The Free Encyclopedia
18 pages
How Webpack Works PDF
No ratings yet
How Webpack Works PDF
26 pages
Bootstrap for ASP.NET MVC - Second Edition
From Everand
Bootstrap for ASP.NET MVC - Second Edition
Pieter van der Westhuizen
5/5 (1)
Aspnet Core Web API Best Practices.03
No ratings yet
Aspnet Core Web API Best Practices.03
30 pages
Solid Principles
100% (1)
Solid Principles
14 pages
C# Programming Solution
No ratings yet
C# Programming Solution
8 pages
Advanced C: Mark Sapossnek CS 594 Computer Science Department Metropolitan College Boston University
No ratings yet
Advanced C: Mark Sapossnek CS 594 Computer Science Department Metropolitan College Boston University
94 pages
Nest - Js A Progressive Node - Js Framework
100% (1)
Nest - Js A Progressive Node - Js Framework
323 pages
React Js
100% (1)
React Js
33 pages
Core With Signal R
No ratings yet
Core With Signal R
3,689 pages
ASP NET Core Succinctly
No ratings yet
ASP NET Core Succinctly
105 pages
Interview Questions and Answers For Freshers PDF
75% (4)
Interview Questions and Answers For Freshers PDF
3 pages
Quick JavaScript Interview Questions - Sandeep Patel
100% (1)
Quick JavaScript Interview Questions - Sandeep Patel
90 pages
Full-Stack Redux Tutorial
No ratings yet
Full-Stack Redux Tutorial
56 pages
Core Web API PDF
No ratings yet
Core Web API PDF
107 pages
Puzzles in Java: Shaping Beginners
From Everand
Puzzles in Java: Shaping Beginners
Kathiravan Udayakumar
1/5 (1)
Mastering Spring Boot 3.0: A comprehensive guide to building scalable and efficient backend systems with Java and Spring
From Everand
Mastering Spring Boot 3.0: A comprehensive guide to building scalable and efficient backend systems with Java and Spring
Ahmet Meric
No ratings yet
Node - Js Hackathon
No ratings yet
Node - Js Hackathon
19 pages
SOLID Principles
No ratings yet
SOLID Principles
8 pages
Mastering JavaScript High Performance - Sample Chapter
No ratings yet
Mastering JavaScript High Performance - Sample Chapter
20 pages
Winforms
No ratings yet
Winforms
48 pages
C# .NET Core
100% (1)
C# .NET Core
707 pages
Design Patterns
100% (1)
Design Patterns
205 pages
Design Patterns Lecture
No ratings yet
Design Patterns Lecture
50 pages
GBD210248 1651 ASM2 Presentation
No ratings yet
GBD210248 1651 ASM2 Presentation
39 pages
Thesis On Software Design Patterns
100% (3)
Thesis On Software Design Patterns
6 pages
Teaching Object-Oriented Concepts Through GUI Programming: January 2007
No ratings yet
Teaching Object-Oriented Concepts Through GUI Programming: January 2007
5 pages
OOAD Syllabus
No ratings yet
OOAD Syllabus
2 pages
Lecture Notes: IV B. Tech I Semester (JNTUH-R13)
No ratings yet
Lecture Notes: IV B. Tech I Semester (JNTUH-R13)
18 pages
Assignment Brief Unit 20 Advanced Programming Sep 2022
No ratings yet
Assignment Brief Unit 20 Advanced Programming Sep 2022
2 pages
Patterson, Hennessy - Computer Organization and Design The Hardware-Software Interface, 2E (Morgan Kaufman, 1997) PDF
No ratings yet
Patterson, Hennessy - Computer Organization and Design The Hardware-Software Interface, 2E (Morgan Kaufman, 1997) PDF
16 pages
04 DPF Simple Extensibility
No ratings yet
04 DPF Simple Extensibility
53 pages
4.gang of Four
No ratings yet
4.gang of Four
4 pages
OOAD Notes
0% (1)
OOAD Notes
28 pages
Olaf Musch - Design Patterns With Java - An Introduction-Springer Vieweg (2023)
No ratings yet
Olaf Musch - Design Patterns With Java - An Introduction-Springer Vieweg (2023)
344 pages
UML Unit-3
No ratings yet
UML Unit-3
10 pages
Design Patterns in The Android Framework
No ratings yet
Design Patterns in The Android Framework
55 pages
Goldsmith Comb
No ratings yet
Goldsmith Comb
146 pages
SOFTWARE SYSTEM DESIGN
No ratings yet
SOFTWARE SYSTEM DESIGN
55 pages
R15 Regulations IV Yr
No ratings yet
R15 Regulations IV Yr
40 pages
Ooad Important Questions
0% (1)
Ooad Important Questions
41 pages
Iptchapter5 151207213412 Lva1 App6891 PDF
No ratings yet
Iptchapter5 151207213412 Lva1 App6891 PDF
17 pages
Seminar Report On CSC 405
No ratings yet
Seminar Report On CSC 405
24 pages
Exam 01
No ratings yet
Exam 01
2 pages
Design Patterns - in Dynamic Languages Presentation
No ratings yet
Design Patterns - in Dynamic Languages Presentation
143 pages
Functional Thinking:: Functional Design Patterns, Part 1
No ratings yet
Functional Thinking:: Functional Design Patterns, Part 1
12 pages
Uid & Ooad Syllabus
No ratings yet
Uid & Ooad Syllabus
2 pages
OOAD
No ratings yet
OOAD
4 pages
Chap 6 - Software Reuse
No ratings yet
Chap 6 - Software Reuse
51 pages
Adapting Software Design Patterns To Develop Reusable IEC 61499 Function Block Applications
No ratings yet
Adapting Software Design Patterns To Develop Reusable IEC 61499 Function Block Applications
9 pages
Design Pattern in C++
No ratings yet
Design Pattern in C++
75 pages

Clean Code in Csharp PDF

Uploaded by

Clean Code in Csharp PDF

Uploaded by

Clean Code in C#

Refactor your legacy C# code base and improve application

Commissioning Editor: Richa Tripathi

First published: July 2020

Production reference: 1170720

Published by Packt Publishing Ltd.

Get a free eBook or video every month

Fully searchable for easy access to vital information

Copy and paste, print, and bookmark content

About the author

Currently employed by a globally recognized leader in the educational software sector, he

Packt is searching for authors like you

Using regions in large classes 31

An example of low coupling 72

Understanding the reasons for a good test 131

Understanding thread safety 238

Further reading 324

Data clump 391

Who this book is for

What this book covers

Chapter 5, Exception Handling, covers checked and unchecked exceptions,

Chapter 11, Addressing Cross-Cutting Concerns, introduces you to using PostSharp to

To get the most out of this book

Visual Studio 2019 Windows 10, macOS

Atom Windows 10, macOS, Linux: https:/​/​atom.​io/​

Azure resources Azure subscription: https:/​/​azure.​microsoft.​com/​en-​gb/​

Azure Key Vault Azure subscription: https:/​/​azure.​microsoft.​com/​en-​gb/​

Obtain your own API key from https:/​/​rapidapi.​com/

Postman Windows 10, macOS, Linux: https:/​/​www.​postman.​com/​

Download the example code files

You can download the code files by following these steps:

1. Log in or register at www.packt.com.

WinRAR/7-Zip for Windows

Download the color images

A block of code is set as follows:

Any command-line input or output is written as follows:

Warnings or important notes appear like this.

Tips and tricks appear like this.

For more information about Packt, please visit packt.com.

The following topics will be covered in this chapter:

The need for coding standards, principles, and methodologies

Understand why bad code negatively impacts projects.

Good code versus bad code

Good Code Bad Code

Incorrectly indented code is also time-consuming to correct, and a frustrating waste of

Comments that state the obvious

Comments that excuse bad code

Here's another example:

Commented-out lines of code

Improper organization of namespaces

MyProject.TextFileMonitor.Core: Core classes that define commonly used

Bad naming conventions

Classes that do multiple jobs

Let's have a look at an example of a class that is doing multiple things:

public void OpenDatabaseConnection() { throw new

#region File Operations

public string ReadText(string filename) { throw new

public void OpenDatabaseConnection() { throw new

public string ReadText(string filename) { throw new

public void WriteFile(string filename, byte[] binaryData) { throw

Methods that do many things

using (CryptoStream cs = new CryptoStream(ms, decryptor,

Methods with more than 10 lines of code

We are now going to take a look at method parameters.

Methods with more than two parameters

Using exceptions to control program flow

The method accepts BusinessRuleException. Depending upon the message in the

In the BetterFlowControlExample() method, a Boolean value is passed into the

Next, we will consider code that is difficult to read.

Code that is difficult to read

Code that is tightly coupled

Here is a bad example of tightly coupled code:

public Database(SqlServerConnection databaseConnection)

Objects left hanging around

Use of the Finalize() method

Learn to Keep It Simple, Stupid

Atom Windows 10, macOS, Linux: https://atom.io/

Azure resources Azure subscription: https://azure.microsoft.com/en-gb/

Azure Key Vault Azure subscription: https://azure.microsoft.com/en-gb/

Obtain your own API key from https://rapidapi.com/

Postman Windows 10, macOS, Linux: https://www.postman.com/