Refactoring with C++: Explore modern ways of developing maintainable and efficient applications

Cover....1

Title Page....2

Copyright and Credits....3

Dedication....4

Contributors....5

Table of Contents....8

Preface....16

Chapter 1: Coding Standards in C++....24

The difference between good code and bad code....24

Why coding standards are important....25

Code convention....26

Language features limitations....28

General guidelines....28

Readability, efficiency, maintainability, and usability....29

Readability....29

Efficiency....30

Maintainability....30

Usability....33

Summary....33

Chapter 2: Main Software Development Principles....34

SOLID....34

The Single Responsibility Principle....35

The Open-Closed Principle....37

The Liskov Substitution Principle....40

The Dependency inversion principle....43

The KISS principle....45

The KISS and SOLID Principles together....46

Side effects and immutability....47

Con.1 – by default, make objects immutable....47

Con.2 – by default, make member functions const....48

Con.3 – by default, pass pointers and references to const....50

Con.4 – use const to define objects with values that do not change after construction....50

Con.5 – use constexpr for values that can be computed at compile time....50

Constness and data races....51

Summary....55

Chapter 3: Causes of Bad Code....56

The need to deliver the product....57

The developer’s personal taste....57

Multiple ways of solving the same problem in C++....58

Revisiting Bob and Alice’s example....58

Raw pointers and C functions versus Standard Library functions....58

Inheritance versus templates....59

Example – handling errors....61

Projects using different approaches....61

Lack of knowledge in C++....62

Using raw pointers and manual memory management....62

Incorrect use of smart pointers....63

Efficient use of move semantics....63

Misusing const correctness....64

Inefficient string handling....65

Undefined behavior with lambdas....65

Misunderstanding undefined behavior....66

Misuse of C-style arrays....66

Insufficient pointer usage....67

Building std::shared_ptr....67

Copying std::shared_ptr by value....67

Cyclic dependencies with std::shared_ptr....68

Checking the std::weak_ptr status....69

Summary....69

Chapter 4: Identifying Ideal Candidates for Rewriting – Patterns and Anti-Patterns....72

What kind of code is worth rewriting?....73

Smelly code and its basic characteristics....73

Anti-patterns....90

The pitfalls of magic numbers – a case study on data chunking....98

Legacy code....100

Summary....104

Chapter 5: The Significance of Naming....106

General naming principles....107

Descriptiveness....107

Consistency....107

Unambiguity....108

Pronounceability....109

Scope and lifetimes....109

Avoid encoding type or scope information....110

Class and method naming....113

Naming variables....114

Utilize namespaces....115

The use of domain-specific language....117

Balancing long names and comments in code....119

Exploring popular C++ coding conventions – Google, LLVM, and Mozilla....120

Summary....121

Chapter 6: Utilizing a Rich Static Type System in C++....122

Utilizing Chrono for time duration....123

Improving Pointer Safety with not_null and std::optional....123

The pitfalls of raw pointers....124

Using not_null from the Guidelines Support Library....124

Utilizing std::optional for optional values....126

A comparison between raw pointers and nullptr....126

Leveraging std::expected for expected results and errors....126

Strong typing with enum class and scoped enumerations....128

A review of enum class....128

The benefits over traditional enums....128

Real-world scenarios....129

Leveraging the standard library’s type utilities....129

std::variant – a type-safe union....129

std::any – type-safe containers for any type....130

Advanced type techniques....131

Avoiding common pitfalls in advanced type usage....133

Writing robust code with type checks....133

Implicit conversions and type coercion....134

Summary....136

Chapter 7: Classes, Objects, and OOP in C++....138

Good candidates for classes....139

Cohesion....139

Encapsulation....139

Reusability....139

Abstraction....139

Real-world entities....139

Manage complexity....140

Minimizing class responsibilities through encapsulation....140

Usage of structs and classes in C++....144

Common method types in classes – getters and setters....146

Inheritance in C++....147

Evolution of inheritance in C++....150

Implementation of inheritance at the binary level....150

Pros and cons of inheritance....150

Base class – Discount....154

Derived class – SeasonalDiscount....155

Derived class – ClearanceDiscount....155

Tight coupling problems....156

Solution – decouple with the strategy pattern....156

Templates and generic programming....160

What are templates good for?....160

Generic algorithms....160

Container classes....161

How templates work....163

How templates are instantiated....164

A real-world example of template usage in C++....165

Defining currencies....165

Defining assets....168

Using the financial system....169

Disadvantages of using templates in system design....169

Summary....176

Chapter 8: Designing and Developing APIs in C++....178

Principles of minimalistic API design....178

Techniques for achieving minimalism....179

Real-world examples of minimalistic API design....182

Common pitfalls and how to avoid them....184

Important caveats of developing shared libraries in C++....184

Shared libraries within a single project....184

Shared libraries for wider distribution....185

Example – MessageSender class....185

Summary....189

Chapter 9: Code Formatting and Naming Conventions....190

Why is code formatting important?....190

Overview of existing tools that facilitate compliance with coding conventions....191

cpplint....192

Artistic Style....193

Uncrustify....193

Editor plugins....194

Clang-Format....195

Clang-Format configuration – a deep dive into customizing your formatting rules....196

Leveraging existing presets....196

Extending and overriding presets....197

Ignoring specific lines with Clang-Format....198

Endless options for configuration....199

Version control and sharing....199

Integrating Clang-Format into the build system....199

Clang-Format report examples....201

Extending for code format checks for CI....205

Clang-Format support across various editors....206

Checking name styling....207

Integrating Clang-Tidy into the build system....209

Checking source code name styling with Clang-Tidy....210

Fixing naming issues automatically....213

Important caveats....214

Example project....215

Clang-Tidy support across various editors....215

Summary....216

Chapter 10: Introduction to Static Analysis in C++....218

The essence of static analysis....218

Leveraging newer compiler versions for enhanced static analysis....219

Compiler settings to harden C++ code....219

GCC....220

Clang....220

MSVC....221

Static analysis via multiple compilers....221

Highlighting compiler differences – unused private members in GCC versus Clang....222

Highlighting compiler differences – compiler checks for uninitialized variables....222

Exploring static analysis with Clang-Tidy....224

Categories of checks in Clang-Tidy....224

Expanding Clang-Tidy’s capabilities with custom checks....225

Fine-tuning Clang-Tidy for customized static analysis....227

Overview of static analysis tools – comparing PVS-Studio, SonarQube, and others to Clang-Tidy....228

PVS-Studio....229

SonarQube....229

Other notable tools....229

Comparison with Clang-Tidy....230

Summary....230

Chapter 11: Dynamic Analysis....232

Compiler-based dynamic code analysis....233

ASan....233

LeakSanitizer (LSan)....253

MemorySanitizer (MSan)....254

TSan....256

UBSan....260

Dynamic code analysis with Valgrind....262

Setting up Valgrind....262

Memcheck – the comprehensive memory debugger....262

Helgrind – threading error detector....264

Performance impact, fine-tuning, and limitations....266

Other notable tools in the Valgrind suite....267

Data Race Detector (DRD) – a thread error detector....267

Cachegrind....267

Callgrind....267

Massif....267

Dynamic heap analysis tool (DHAT)....267

Summary....268

Chapter 12: Testing....270

Test-driven development....271

Unit testing in C++....272

C++ unit testing frameworks....272

Google Test and Google Mock....273

Integrating Google Test into a C++ project....273

Usage of Google Test in C++ projects....275

Writing a simple test....275

Using a test fixture....276

The main function....277

Running Google Test tests....277

Advanced features of Google Test....279

Using gMock in C++ projects....280

Example of using gMock....281

Mocking non-virtual methods via dependency injection....284

Mocking with templates....286

The Nice, the Strict, and the Naggy....288

Other notable C++ unit testing frameworks....292

Catch2....292

Boost.Test....292

Doctest....292

Google Test versus Catch2 versus Boost.Test versus Doctest....292

Good candidates for unit tests....293

E2E testing in software development....294

E2E testing frameworks....294

When to use E2E testing....294

Situations favoring E2E testing....295

Complex interactions....295

Real-world environment testing....296

Automatic test coverage tracking tools....297

Automatic test coverage tracking tools with examples....297

Utilizing hit maps for enhanced test coverage analysis....298

Recommendations for code coverage....300

Summary....301

Chapter 13: Modern Approach to Managing Third Parties....302

Overview of linking and shared V threads::ThreadsS static libraries....303

Managing third-party libraries in C++....304

Installing libraries with OS package managers....304

Using Git as a third-party manager via submodules....305

Using CMake FetchContent to download libraries....305

Conan – advanced dependency management....306

Conan configuration and features....306

Library locations and Conan Center....306

Configuring static or dynamic linking....307

Extending Conan with custom packages....307

CMake integration....308

Other build system integration....308

Custom integration....309

Conclusion....309

vcpkg....309

Key differences from Conan....309

Operating system support....310

Example of configuring a project with vcpkg....310

Utilizing Docker for C++ builds....311

Summary....319

Chapter 14: Version Control....322

What is a good commit?....323

The principle of singular focus....323

The art of communication....323

The art of refinement....324

Conventional Commits specification....325

Linking code to context....325

Overview and intent....326

Options and usage....326

Origins and adoption....328

Advantages of Conventional Commits....328

Commitlint – enforcing commit message standards....328

Installation....329

Configuration....329

Local usage....329

Customizing rules....330

Basic configuration....330

Custom rule configuration....331

Scope and subject configuration....331

Customizing and sharing configurations....332

Integration with CI....332

Generating changelogs....335

Installation....335

GitCliff usage....335

Utilizing git-bisect in bug hunting....339

Summary....342

Chapter 15: Code Review....344

What is a code review and why is it needed?....344

Benefits of code reviews....345

Preparing for code reviews....346

Clear guidelines....346

Self-review....347

How to pass a code review....348

Discuss big features with reviewers and code owners before writing code....348

Go over your code before publishing it....348

Make sure the code is compliant with the code convention....348

Code review is a conversation, not an order....348

Remember – your code is not you....349

How to efficiently dispute during a code review....349

Clear justification for changes....349

Reciprocal explanation from reviewees....349

Direct communication....350

Involving additional perspectives....350

How to be a good reviewer....350

Initiate the conversation....350

Maintain politeness and respect....350

Review manageable chunks....351

Avoid personal bias....351

Focus on understandability....351

Summary....352

Index....354

About Packt....364

Other Books You May Enjoy....365

Improve readability and understandability of code using C++ best practices

Key Features

• Enrich your coding skills using features from the modern C++ standard and industry approved libraries
• Implement refactoring techniques and SOLID principles in C++
• Apply automated tools to improve your code quality
• Purchase of the print or Kindle book includes a free PDF eBook

Despite the prevalence of higher-level languages, C++ is still running the world, from bare-metal embedded systems to distributed cloud-native systems. C++ is on the frontline whenever there is a need for a performance-sensitive tool supporting complex data structures. The language has been actively evolving for the last two decades.

This book is a comprehensive guide that shows you how to implement SOLID principles and refactor legacy code using the modern features and approaches of C++, the standard library, Boost library collection, and Guidelines Support Library by Microsoft. The book begins by describing the essential elements of writing clean code and discussing object-oriented programming in C++. You'll explore the design principles of software testing with examples of using popular unit testing frameworks such as Google Test. The book also guides you through applying automated tools for static and dynamic code analysis using Clang Tools.

By the end of this book, you'll be proficient in applying industry-approved coding practices to design clean, sustainable, and readable real-world C++ code.

What You Will Learn:

• Leverage the rich type system of C++ to write safe and elegant code
• Create advanced object-oriented designs using the unique features of C++
• Minimize code duplication by using metaprogramming
• Refactor code safely with the help of unit tests
• Ensure code conventions and format with clang-format
• Facilitate the usage of modern features automatically with clang-tidy
• Catch complex bugs such as memory leakage and data races with Clang AddressSanitizer and ThreadSanitizer

Who this book is for:

This book will benefit experienced C++ programmers the most, but is also suitable for technical leaders, software architects, and senior software engineers who want to save on costs and improve software development process efficiency by using modern C++ features and automated tools.