std::floating_point (since C++20)

From cppreference.com
< cpp‎ | concepts
Defined in header <concepts>
template< class T >
concept floating_point = std::is_floating_point_v<T>;
(since C++20)

The concept floating_point<T> is satisfied if and only if T is a floating-point type.

Example

#include <concepts>
#include <iostream>
#include <type_traits>
 
constexpr std::floating_point auto x2(std::floating_point auto x)
{
    return x + x;
}
 
constexpr std::integral auto x2(std::integral auto x)
{
    return x << 1;
}
 
int main()
{
    constexpr auto d = x2(1.1);
    static_assert(std::is_same_v<double const, decltype(d)>);
    std::cout << d << '\n';
 
    constexpr auto f = x2(2.2f);
    static_assert(std::is_same_v<float const, decltype(f)>);
    std::cout << f << '\n';
 
    constexpr auto i = x2(444);
    static_assert(std::is_same_v<int const, decltype(i)>);
    std::cout << i << '\n';
}

Output:

2.2
4.4
888

References

  • C++23 standard (ISO/IEC 14882:2024):
  • 18.4.7 Arithmetic concepts [concepts.arithmetic]
  • C++20 standard (ISO/IEC 14882:2020):
  • 18.4.7 Arithmetic concepts [concepts.arithmetic]

See also

checks if a type is a floating-point type
(class template)