C++ Unleashed: From Zero to Hero

Previous chapter: Templates and Generic Programming

Go to Table of Contents

Chapter 9: Modern C++ Features

Modern C++ (C++11 and later) introduced a range of features that enhance the language’s expressiveness, performance, and safety. This chapter covers some of the most important features that you should know to write effective and modern C++ code today.

Table of Contents

1. Type Inference with auto
2. Range-Based For Loops
3. Type Aliases (using)
4. constexpr, consteval, and constinit
   • Understanding constexpr
   • Introducing consteval
   • Exploring constinit
5. decltype
6. std::variant and std::optional
7. nullptr

---

Type Inference with auto

Introduction

The auto keyword allows the compiler to automatically deduce the type of a variable at compile time. This feature simplifies code and improves readability, especially when dealing with complex types.

Usage

Basic Example:

#include <iostream>
#include <vector>

int main() {
    auto x = 42; // x is int
    auto y = 3.14; // y is double
    auto str = "Hello, World!"; // str is const char*

    std::vector<int> vec = {1, 2, 3, 4, 5};
    auto it = vec.begin(); // it is std::vector<int>::iterator

    std::cout << "x: " << x << ", y: " << y << ", str: " << str << std::endl;

    return 0;
}

Output:

x: 42, y: 3.14, str: Hello, World!

Best Practices

• Use auto When Type is Obvious:

  auto count = 10; // Clearly an int
  

• Avoid Obscuring Types:

  auto result = someFunction(); // If the return type of someFunction is unclear
  

• Combine with decltype for Complex Deductions:

  decltype(auto) ref = someFunctionReturningReference();
  

• Prefer Explicit Types for Function Return Types and Parameters:

  int add(int a, int b); // Clear return type
  

---

Range-Based For Loops

Introduction

Range-based for loops provide a more readable and concise way to iterate through containers, eliminating the need for explicit iterators or index-based loops. Introduced in C++11, they enhance code clarity and reduce the likelihood of errors associated with traditional loop constructs.

Syntax

for (auto& element : container) {
    // Use element
}

Example

#include <iostream>
#include <vector>

int main() {
    std::vector<int> numbers = {1, 2, 3, 4, 5};

    for (const auto& num : numbers) {
        std::cout << num << " "; // Outputs: 1 2 3 4 5
    }
    std::cout << std::endl;

    return 0;
}

Output:

1 2 3 4 5

Best Practices

• Use const auto& When Not Modifying Elements:

  for (const auto& num : numbers) {
      // Read-only access
  }
  

• Use auto& When Modifying Elements:

  for (auto& num : numbers) {
      num *= 2; // Modify elements
  }
  

• Avoid Unnecessary Copies:

  for (const auto& str : stringList) { /* ... */ } // Avoid copying strings
  

• Use auto&& for Perfect Forwarding:

  for (auto&& element : container) { /* ... */ }
  

---

Type Aliases (using)

Introduction

Type aliases allow you to create alternative names for existing types, enhancing code readability and maintainability. The using keyword is preferred over typedef in modern C++ due to its clearer syntax and enhanced capabilities, especially with templates.

Syntax

using AliasName = ExistingType;

Example

#include <vector>
#include <string>

using StringList = std::vector<std::string>;

int main() {
    StringList names = {"Alice", "Bob", "Charlie"};
    return 0;
}

Best Practices

• Simplify Complex Type Declarations:

  using IntPair = std::pair<int, int>;
  

• Enhance Code Readability:

  using Byte = unsigned char;
  

• Use Descriptive Alias Names:

  using Matrix = std::vector<std::vector<double>>;
  

• Leverage with Templates:

  template <typename T>
  using Vec = std::vector<T>;
  
  Vec<int> intVec;
  

---

constexpr, consteval, and constinit

Modern C++ introduces several keywords that enhance compile-time programming, enabling more efficient and safer code. Understanding the distinctions and appropriate usage of constexpr, consteval, and constinit is crucial for leveraging these features effectively.

Understanding constexpr

constexpr specifies that the value of a variable or the return value of a function can be evaluated at compile time. It allows for optimizations and ensures that certain computations are performed during compilation rather than at runtime.

Usage:

#include <iostream>

constexpr int square(int x) {
    return x * x;
}

int main() {
    constexpr int result = square(5); // Computed at compile time
    int runtime_result = square(6);   // Also allowed, computed at runtime
    std::cout << "Square of 5 is: " << result << std::endl;
    std::cout << "Square of 6 is: " << runtime_result << std::endl;
    return 0;
}

Output:

Square of 5 is: 25
Square of 6 is: 36

Key Points:

• Functions and Variables: Both can be declared constexpr.
• Compile-Time Evaluation: When used in a context that requires a compile-time constant, the evaluation must occur at compile time.
• Flexibility: constexpr functions can be used in both compile-time and runtime contexts.

Introducing consteval

consteval is a keyword introduced in C++20 that declares an immediate function. These functions are guaranteed to produce a compile-time constant and must be evaluated at compile time. Unlike constexpr, which allows both compile-time and runtime evaluations, consteval enforces compile-time evaluation.

Usage:

#include <iostream>

consteval int square(int x) {
    return x * x;
}

int main() {
    constexpr int sq = square(5); // Valid
    // int runtime_sq = square(5); // Error: requires compile-time evaluation
    std::cout << "Square of 5 is: " << sq << std::endl;
    return 0;
}

Output:

Square of 5 is: 25

Key Points:

• Immediate Evaluation: Must be evaluated at compile time.
• Type Safety: Ensures that functions producing constants are not mistakenly used in runtime contexts.
• Use Cases: Ideal for functions that must generate constant expressions, enhancing both performance and reliability.

Exploring constinit

constinit is another keyword introduced in C++20 that ensures a variable is constantly initialized at compile time without making it immutable. Unlike constexpr, which makes a variable a compile-time constant, constinit only guarantees that the variable is initialized at compile time, allowing it to remain mutable during runtime.

Usage:

#include <iostream>

constinit int global_counter = 0;

int main() {
    global_counter += 5; // Allowed, since constinit doesn't make it immutable
    std::cout << "Global Counter: " << global_counter << std::endl;
    return 0;
}

Output:

Global Counter: 5

Key Points:

• Static Initialization: Ensures that variables are initialized before any dynamic initialization occurs, preventing the static initialization order fiasco.
• Mutable Variables: Unlike constexpr, variables declared with constinit can be modified after initialization.
• Thread Safety: Guarantees that variables are initialized in a thread-safe manner at compile time.
• Use Cases: Ideal for global and static variables that require early initialization but need to remain mutable during runtime.

Best Practices

• Use constexpr: When you want flexibility for compile-time and runtime evaluations. Ideal for functions and variables that can benefit from both contexts.

• Use consteval: When you need to guarantee that a function is evaluated at compile time, ensuring compile-time constants and preventing runtime evaluations.

• Use constinit: For variables that must be initialized at compile time but need to remain mutable. Essential for global and static variables to ensure proper initialization order.

---

decltype

Introduction

The decltype keyword in C++ deduces the type of an expression at compile time. It is particularly useful in templates and when the type of an expression is complex or not easily determined. decltype enhances type safety and flexibility, allowing for more generic and adaptable code.

Usage

Basic Example:

#include <iostream>

int main() {
    int x = 5;
    decltype(x) y = 10; // y is of type int
    std::cout << "y: " << y << std::endl;
    return 0;
}

Output:

y: 10

With Complex Expressions:

#include <vector>
#include <iostream>

int main() {
    std::vector<int> vec = {1, 2, 3, 4, 5};
    decltype(vec.begin()) it = vec.begin(); // it is std::vector<int>::iterator

    std::cout << "First element: " << *it << std::endl;
    return 0;
}

Output:

First element: 1

Best Practices

• Use in Templates: Facilitate generic programming by deducing return types or variable types based on template parameters.

• Combine with auto: Create concise type declarations without sacrificing clarity.

  auto x = someFunction();
  decltype(x) y = anotherFunction();
  

• Avoid Overuse: While powerful, excessive use of decltype can make code harder to read. Use it judiciously to maintain clarity.

---

std::variant and std::optional

Introduction

C++17 introduced std::variant and std::optional to handle situations where a value might belong to multiple types or might be absent, respectively. These utilities enhance type safety and expressiveness in your code, reducing the reliance on traditional approaches like unions or sentinel values.

std::variant

Description:

std::variant is a type-safe union that can hold one value from a specified set of types. It provides a way to store and manage multiple types in a single variable, ensuring that only one type is active at any given time.

Usage:

#include <variant>
#include <iostream>
#include <string>

int main() {
    std::variant<int, std::string> data;
    
    data = 10;
    std::cout << std::get<int>(data) << std::endl; // Outputs: 10
    
    data = "Hello, Variant!";
    std::cout << std::get<std::string>(data) << std::endl; // Outputs: Hello, Variant!
    
    return 0;
}

Handling Multiple Types with std::visit:

#include <variant>
#include <iostream>
#include <string>

struct Visitor {
    void operator()(int i) const { std::cout << "int: " << i << std::endl; }
    void operator()(const std::string& str) const { std::cout << "string: " << str << std::endl; }
};

int main() {
    std::variant<int, std::string> data = "Hello";
    std::visit(Visitor{}, data); // Outputs: string: Hello
    
    data = 42;
    std::visit(Visitor{}, data); // Outputs: int: 42
    
    return 0;
}

Output:

string: Hello
int: 42

Best Practices

• Use std::variant for Heterogeneous Data: When a variable needs to hold one of several types, std::variant provides a type-safe alternative to unions.

• Handle All Possible Types: Use std::visit to ensure that all possible types are handled, enhancing type safety.

• Avoid Excessive Nesting: Deeply nested std::variant instances can complicate code. Consider refactoring if necessary.

• Leverage with std::monostate: Use std::monostate as the first type in a std::variant to represent an empty state.

std::optional

Description:

std::optional represents an object that may or may not contain a value, eliminating the need for sentinel values like nullptr or special error codes. It enhances code clarity by explicitly indicating the presence or absence of a value.

Usage:

#include <optional>
#include <iostream>

std::optional<int> findValue(bool shouldFind) {
    if (shouldFind) {
        return 42; // Optional value is present
    }
    return std::nullopt; // No value present
}

int main() {
    auto value = findValue(true);
    if (value) {
        std::cout << "Found value: " << *value << std::endl; // Dereference
    } else {
        std::cout << "No value found." << std::endl;
    }

    return 0;
}

Output:

Found value: 42

Simplified Access with operator* and operator->:

#include <optional>
#include <iostream>
#include <string>

std::optional<std::string> getName(bool valid) {
    if (valid)
        return "Alice";
    else
        return std::nullopt;
}

int main() {
    auto nameOpt = getName(true);
    if (nameOpt) {
        std::cout << "Name: " << *nameOpt << std::endl; // Outputs: Name: Alice
    }

    return 0;
}

Output:

Name: Alice

Best Practices

• Use std::optional for Optional Return Values: Functions that might not always return a meaningful value can return std::optional instead of using sentinel values.

• Avoid Using std::optional for Class Members: Prefer other design patterns, such as the Null Object pattern, to handle optional class members.

• Prefer Checking has_value() or Using if (opt): Ensure that the presence of a value is checked before accessing it to prevent exceptions.

• Leverage value_or for Default Values: Provide a fallback value when the optional is empty.

  int value = opt.value_or(0);
  

---

nullptr

Introduction

Introduced in C++11, nullptr is a type-safe pointer literal representing a null pointer. It replaces the traditional NULL macro and integer literal 0, providing better type safety and clarity in your code.

Usage

Basic Example:

#include <iostream>

void func(int* ptr) {
    if (ptr == nullptr) {
        std::cout << "Pointer is null." << std::endl;
    }
}

int main() {
    int* p = nullptr; // Safe null pointer
    func(p);

    return 0;
}

Output:

Pointer is null.

Function Overloading Ambiguity Resolution:

#include <iostream>

void func(int) {
    std::cout << "Function with int parameter." << std::endl;
}

void func(char*) {
    std::cout << "Function with char* parameter." << std::endl;
}

int main() {
    func(nullptr); // Calls func(char*) due to type safety
    return 0;
}

Output:

Function with char* parameter.

Advantages Over NULL and 0

• Type Safety: nullptr has its own type (std::nullptr_t), preventing unintended conversions and function overloading ambiguities.

• Clarity: Clearly indicates that the pointer is intended to be null, improving code readability.

• Avoids Ambiguity: Eliminates confusion between integer 0 and null pointers in function overloading scenarios.

Best Practices

• Use nullptr Instead of NULL or 0:

  int* ptr = nullptr;
  

• Consistently Use nullptr Across Codebase: Ensure uniformity to leverage its type safety benefits fully.

• Avoid Mixing nullptr with Other Null Representations: Stick to nullptr to maintain consistency and clarity.

---

Summary

In this chapter, you learned about important features introduced in modern C++:

• Type Inference with auto: Automatically deducing variable types.
• Range-Based For Loops: Simplifying iterations over containers.
• Type Aliases (using): Creating alternative names for existing types.
• constexpr, consteval, and constinit: Enhancing compile-time computations and initialization.
• decltype: Deducing the type of expressions at compile time.
• std::variant and std::optional: Handling multiple types and optional values safely.
• nullptr: A type-safe null pointer constant.

These features make modern C++ more powerful, expressive, and safer, helping you write better code with less risk of errors. In the next chapter, we will delve into the Standard Template Library (STL), exploring its containers, algorithms, and iterators.

Next chapter: Standard Template Library (STL)