Coming in C++20

The main features in C++20 of interest to scientific computing are:

Modules

Concepts

Views

C++ 20 introduces the ranges library. This allows us to more easily consider views into our containers.

Here’s an example of using a range-based for loop over a set of integers:

Listing 139 iota_loop.cpp

#include <iostream>
#include <ranges>

int main() {

    for (auto i : std::views::iota(2, 7)) {
        std::cout << i << std::endl;
    }

}

Range Adaptors

Range adaptors look like pipes that we saw when discussing Bash (see: https://learn.microsoft.com/en-us/cpp/standard-library/range-adaptors?view=msvc-170)

Here’s an example of using an adaptor to reverse the iteration through a vector using a range-based for loop:

Listing 140 reverse_adaptor.cpp

#include <iostream>
#include <vector>
#include <ranges>

int main() {

    std::vector<double> v{1, 2, 3, 4, 5};

    for (auto e : v | std::views::reverse) {
        std::cout << e << " ";
    }
    std::cout << std::endl;

}

3-way Comparison

<numbers>

The numbers header provides mathematical constants. They are implemented as templates that can be defined with whatever type you need:

Listing 141 numbers_example.cpp

#include <iostream>
#include <iomanip>
#include <limits>
#include <numbers>

int main() {

    std::cout << std::setprecision(std::numeric_limits<float>::digits10+1)
              << std::numbers::e_v<float> << std::endl;
    std::cout << std::setprecision(std::numeric_limits<double>::digits10+1)
              << std::numbers::e_v<double> << std::endl;
    std::cout << std::setprecision(std::numeric_limits<long double>::digits10+1)
              << std::numbers::e_v<long double> << std::endl;

}

<format>

We already saw the new style formatting when we looked at std::print(). The format is handled by the format header.