C++ std:any Examples

[Linux Hint]

In C++ programming, “std::any” from the Standard Template Library (STL) introduces a dynamic typing to handle a heterogeneous data. Unlike traditional containers, “std::any” allows storing the values of any type within a single container, enhancing the flexibility in scenarios where the data types are unknown or vary at runtime. This type-agnostic approach promotes a generic programming which empowers the developers to create a more adaptable and expressive code while maintaining the type safety. In this exploration, we’ll delve into the features of “std::any”, its usage patterns, and practical examples that illustrate its role in writing a robust and flexible C++ code.

Example 1: Basic Usage of Std::Any

First, let’s explore a straightforward example to demonstrate the fundamental usage of “std::any”. Consider a scenario where you need a function to accept various types of parameters:

Here is the code snippet:

#include <iostream>
#include <any>

void processAny(const std::any& value) {
    if (value.has_value()) {
        std::cout << "Type of stored value: " << value.type().name() << std::endl;

        if (value.type() == typeid(int)) {
            std::cout << "Value: " << std::any_cast<int>(value) << std::endl;
        } else if (value.type() == typeid(double)) {
            std::cout << "Value: " << std::any_cast<double>(value) << std::endl;
        } else if (value.type() == typeid(std::string)) {
            std::cout << "Value: " << std::any_cast<std::string>(value) << std::endl;
        } else {
            std::cout << "Unsupported type!" << std::endl;
        }
    } else {
        std::cout << "No value stored in std::any." << std::endl;
    }
}

int main() {
    processAny(42);
    processAny(3.14);
    processAny(std::string("Hello, std::any!"));
    processAny(4.5f); // Unsupported type

    return 0;
}

 
In this example, we define the “processAny” function that takes an “std::any” reference as a parameter and examines its content. Inside the function, we first check if the “std::any” variable has a stored value using has_value(). If a value is present, we determine the type of the stored value using type().name() and proceed to print the corresponding value based on its type. The main function then demonstrates the utility of “processAny” by calling it with different types: an integer (42), a double (3.14), and a string (“Hello, std::any!”). The function appropriately handles each type and prints the respective values. However, when attempting to process a floating-point number (4.5f), which is unsupported in this example, the program gracefully handles the situation by indicating that the type is unsupported.

The generated output is:

This showcases how “std::any” enables the dynamic handling of various data types, making it a versatile tool for generic programming in C++.

Example 2: Storing the User-Defined Types

The second example explores how this dynamic type within the Standard Template Library (STL) seamlessly accommodates the custom data structures. Focusing on a user-defined type, the point structure, we showcase how “std::any” handles the instances of such structures.

Here is the code:

#include <iostream>
#include <any>

class MyClass {
public:
    MyClass(int value) : data(value) {}

    void printData() const {
        std::cout << "Data in MyClass: " << data << std::endl;
    }

private:
    int data;
};

int main() {
    std::any anyObject = MyClass(42);

    if (anyObject.has_value()) {
        auto& myClassInstance = std::any_cast<MyClass&>(anyObject);
        myClassInstance.printData();
    } else {
        std::cout << "No value stored in std::any." << std::endl;
    }

    return 0;
}

 
In this C++ code snippet, we create a simple example to illustrate using the “std::any” type with a user-defined class called “MyClass”. Within the class, there’s a private member variable called “data” and a public method called printData() to display the value of this data. An integer value is passed and assigned to the “data” member in the constructor.

In the “main” function, we instantiate an object of “MyClass” with an initial value of 42 and then store it in the “std::any” variable named “anyObject”. This demonstrates the capability of “std::any” to hold the instances of user-defined classes.

Following this, we use an “if” statement to check whether “anyObject” has a value using the has_value() method. If there is a value, we retrieve the stored object using “std::any_cast”. The “std::any_cast” is employed with the “MyClass&” template argument to cast the stored object to a reference of “MyClass”. This reference, “myClassInstance”, is then used to call the printData() method, showcasing the ability to access and operate on the stored user-defined type within the “std::any”.

If no value is stored in “std::any”, we print a message signifying this. This conditional check ensures that we handle the scenarios where the “std::any” variable might be empty.

Here is the output:

Example 3: Container of Mixed Types

In programming, a “mixed-type container” refers to a data structure that is capable of holding the elements of diverse, potentially unrelated data types. This flexibility is valuable when dealing with scenarios where the data types are unknown at compile time or dynamically changing during program execution. In C++, “std::any” exemplifies this concept, allowing the creation of a single container to store the values of different types.

Let’s explore a scenario where we create a container that holds various types:

#include <iostream>
#include <any>
#include <vector>

int main() {

    std::vector<std::any> mixedContainer;

    mixedContainer.push_back(42);                      
    mixedContainer.push_back(3.14);                    
    mixedContainer.push_back(std::string("Hello"));  
    mixedContainer.push_back(true);                    

    for (const auto& element : mixedContainer) {
        if (element.type() == typeid(int)) {
            std::cout << "Integer: " << std::any_cast<int>(element) << std::endl;
        } else if (element.type() == typeid(double)) {
            std::cout << "Double: " << std::any_cast<double>(element) << std::endl;
        } else if (element.type() == typeid(std::string)) {
            std::cout << "String: " << std::any_cast<std::string>(element) << std::endl;
        } else if (element.type() == typeid(bool)) {
            std::cout << "Boolean: " << std::any_cast<bool>(element) << std::endl;
        } else {
            std::cout << "Unknown type" << std::endl;
        }
    }

    return 0;
}

 
In this illustration, we demonstrate the concept of a mixed-type container using C++ and the “std::any” feature. We create the “std::vector<std::any>” named “mixedContainer” to serve as our container to hold the elements of different data types. Using the “push_back” function, we populate this container with various elements including an integer (42), a double (3.14), a string (“Hello”), and a Boolean (true).

As we iterate through the “mixedContainer” using a “for” loop, we employ the type() function to identify each element’s data type dynamically. Utilizing “std::any_cast”, we extract and print the corresponding values based on their types. For instance, if the element is of type “int”, we print it as an integer. If it’s of type “double”, we print it as a double, and so on.

Here is the generated output:

Example 4: Error Handling with Std::Any

Handling errors when using “std::any” involves checking whether the type is supported or whether a value is stored. In this example, we demonstrate how to handle the unsupported types:

#include <iostream>
#include <any>

int main() {
    std::any myAny = 42;

    try {
   
        double value = std::any_cast<double>(myAny);
        std::cout << "Value: " << value << std::endl;
    } catch (const std::bad_any_cast& e) {
       
        std::cerr << "Error: " << e.what() << std::endl;
    }

    return 0;
}

 
We start by initializing the “std::any” variable, “myAny”, with the value of 42 of the integer type. Inside the subsequent “try” block, we make an explicit attempt to cast this integer value into a “double” using the “std::any_cast<double>” operation. However, since the actual type that is stored in “myAny” is an integer, this casting operation is invalid for a “double” which leads to a mismatch type.

To manage this potential error gracefully, we implement the exception handling with a “catch” block that is designed to catch the specific exception type of “std::bad_any_cast”. In the event of an unsuccessful cast, the “catch” block is activated and we generate an error message using “std::cerr” to communicate the nature of the error. This error-handling strategy ensures that our program can gracefully handle the situations where the attempted type cast clashes with the actual type that is stored in the “std::any” variable.

Conclusion

In this article, we explored the applications of “std::any” in C++, a dynamic type container that is introduced in C++ for values of diverse types. We demonstrated its versatility through various examples, showcasing scenarios that range from basic usage to handling the user-defined types and heterogeneous collections. We demonstrated its practical application in scenarios where the type of data is not known at compile time. Additionally, we explored about the error-handling techniques, emphasizing the importance of gracefully managing the unsupported types through exception handling.

View the full article