W3cubDocs

/C++

std::cmp_equal, cmp_not_equal, cmp_less, cmp_greater, cmp_less_equal, cmp_greater_equal

Defined in header <utility>
template< class T, class U >
constexpr bool cmp_equal( T t, U u ) noexcept;
(1) (since C++20)
template< class T, class U >
constexpr bool cmp_not_equal( T t, U u ) noexcept;
(2) (since C++20)
template< class T, class U >
constexpr bool cmp_less( T t, U u ) noexcept;
(3) (since C++20)
template< class T, class U >
constexpr bool cmp_greater( T t, U u ) noexcept;
(4) (since C++20)
template< class T, class U >
constexpr bool cmp_less_equal( T t, U u ) noexcept;
(5) (since C++20)
template< class T, class U >
constexpr bool cmp_greater_equal( T t, U u ) noexcept;
(6) (since C++20)

Compare the values of two integers t and u. Unlike builtin comparison operators, negative signed integers always compare less than (and not equal to) unsigned integers: the comparison is safe against lossy integer conversion.

-1 > 0u; // true
std::cmp_greater(-1, 0u); // false

It is a compile-time error if either T or U is not a signed or unsigned integer type (including standard integer type and extended integer type).

Parameters

t - left-hand argument
u - right-hand argument

Return value

1) true if t is equal to u.
2) true if t is not equal to u.
3) true if t is less than u.
4) true if t is greater than u.
5) true if t is less or equal to u.
6) true if t is greater or equal to u.

Possible implementation

template<class T, class U>
constexpr bool cmp_equal(T t, U u) noexcept
{
    using UT = std::make_unsigned_t<T>;
    using UU = std::make_unsigned_t<U>;
    if constexpr (std::is_signed_v<T> == std::is_signed_v<U>)
        return t == u;
    else if constexpr (std::is_signed_v<T>)
        return t < 0 ? false : UT(t) == u;
    else
        return u < 0 ? false : t == UU(u);
}
 
template<class T, class U>
constexpr bool cmp_not_equal(T t, U u) noexcept
{
    return !cmp_equal(t, u);
}
 
template<class T, class U>
constexpr bool cmp_less(T t, U u) noexcept
{
    using UT = std::make_unsigned_t<T>;
    using UU = std::make_unsigned_t<U>;
    if constexpr (std::is_signed_v<T> == std::is_signed_v<U>)
        return t < u;
    else if constexpr (std::is_signed_v<T>)
        return t < 0 ? true : UT(t) < u;
    else
        return u < 0 ? false : t < UU(u);
}
 
template<class T, class U>
constexpr bool cmp_greater(T t, U u) noexcept
{
    return cmp_less(u, t);
}
 
template<class T, class U>
constexpr bool cmp_less_equal(T t, U u) noexcept
{
    return !cmp_greater(t, u);
}
 
template<class T, class U>
constexpr bool cmp_greater_equal(T t, U u) noexcept
{
    return !cmp_less(t, u);
}

Notes

These functions cannot be used to compare enums (including std::byte), char, char8_t, char16_t, char32_t, wchar_t and bool.

Feature-test macro Value Std Comment
__cpp_lib_integer_comparison_functions 202002L (C++20) Integer comparison functions

Example

The example below might produce different signedness comparison warning if compiled without an appropriate warning suppression flag, e.g., -Wno-sign-compare (gcc/clang with -Wall -Wextra, see also SO: disabling a specific warning).

#include <utility>
 
// Uncommenting the next line will disable "signed/unsigned comparison" warnings:
// #pragma GCC diagnostic ignored "-Wsign-compare"
 
int main()
{
    static_assert(sizeof(int) == 4); // precondition
 
    // Quite surprisingly
    static_assert(-1 > 1U); //< warning: sign-unsign comparison
    // because after implicit conversion of -1 to the RHS type (`unsigned int`)
    // the expression is equivalent to:
    static_assert(0xFFFFFFFFU > 1U);
    static_assert(0xFFFFFFFFU == static_cast<unsigned>(-1));
 
    // In contrast, the cmp_* family compares integers as most expected -
    // negative signed integers always compare less than unsigned integers:
    static_assert(std::cmp_less(-1, 1U));
    static_assert(std::cmp_less_equal(-1, 1U));
    static_assert(!std::cmp_greater(-1, 1U));
    static_assert(!std::cmp_greater_equal(-1, 1U));
 
    static_assert(-1 == 0xFFFFFFFFU); //< warning: sign-unsign comparison
    static_assert(std::cmp_not_equal(-1, 0xFFFFFFFFU));
}

See also

function object implementing x == y
(class template)
function object implementing x != y
(class template)
function object implementing x < y
(class template)
function object implementing x > y
(class template)
function object implementing x <= y
(class template)
function object implementing x >= y
(class template)
(C++20)
function object implementing x == y
(class)
(C++20)
function object implementing x != y
(class)
(C++20)
function object implementing x < y
(class)
(C++20)
function object implementing x > y
(class)
(C++20)
function object implementing x <= y
(class)
(C++20)
function object implementing x >= y
(class)
(C++20)
function object implementing x <=> y
(class)
(C++20)
checks if an integer value is in the range of a given integer type
(function template)
provides an interface to query properties of all fundamental numeric types.
(class template)

© cppreference.com
Licensed under the Creative Commons Attribution-ShareAlike Unported License v3.0.
https://en.cppreference.com/w/cpp/utility/intcmp