rs-core

Core utilities

---

Project maintained by CaptainCrowbar Hosted on GitHub Pages — Theme by mattgraham

128-Bit Integers

Core utility library by Ross Smith

#include "rs-core/uint128.hpp"
namespace RS;

128-bit integers

class Uint128;
using uint128_t = [see below];

The Uint128 class implements 128-bit arithmetic using a pair of 64-bit integers internally, and is described in detail below. The uint128_t type is defined as a convenience alias for the native 128-bit type if one exists, otherwise for the Uint128 class.

using Uint128::word = std::uint64_t;

The internal word size.

constexpr Uint128::Uint128() noexcept;

The default constructor sets the value to zero.

template <std::integral T> constexpr [conditionally explicit]
    Uint128::Uint128(T t) noexcept;

Conversion constructor from any primitive integer type. This is implicit if the argument type is unsigned and no larger than 128 bits, otherwise explicit. If T is signed, negative arguments will be represented using twos complement.

constexpr Uint128::Uint128(std::uint64_t hi, std::uint64_t lo) noexcept;

Constructor from the high and low 64-bit halves.

explicit Uint128::Uint128(std::string_view str, int base = 10);

Constructor from a string. This will throw std::out_of_range if the base is out of range (2-36), or std:invalid_argument if the string is empty or contains a character that is not a valid digit in the specified base. Values out of range will be truncated to their low 128 bits.

constexpr Uint128::~Uint128() noexcept;
constexpr Uint128::Uint128(const Uint128& u) noexcept;
constexpr Uint128::Uint128(Uint128&& u) noexcept;
constexpr Uint128& Uint128::operator=(const Uint128& u) noexcept;
constexpr Uint128& Uint128::operator=(Uint128&& u) noexcept;

Other life cycle functions.

explicit constexpr Uint128::operator bool() const noexcept;

Explicit boolean conversion operator. As usual, this returns true if the value is not zero.

template <std::integral T>
    constexpr explicit Uint128::operator T() const noexcept;

Explicit conversion operator to a primitive integer type. For unsigned types, the value will be truncated to the number of bits in the return type; for signed types, behaviour is undefined if the value does not fit in the return type.

constexpr Uint128& Uint128::operator++() noexcept;
constexpr Uint128 Uint128::operator++(int) noexcept;
constexpr Uint128& Uint128::operator--() noexcept;
constexpr Uint128 Uint128::operator--(int) noexcept;
constexpr Uint128& Uint128::operator+=(Uint128 y) noexcept;
constexpr Uint128& Uint128::operator-=(Uint128 y) noexcept;
constexpr Uint128& Uint128::operator*=(Uint128 y) noexcept;
constexpr Uint128& Uint128::operator/=(Uint128 y) noexcept;
constexpr Uint128& Uint128::operator%=(Uint128 y) noexcept;
constexpr Uint128 operator+(Uint128 x, Uint128 y) noexcept;
constexpr Uint128 operator-(Uint128 x, Uint128 y) noexcept;
constexpr Uint128 operator*(Uint128 x, Uint128 y) noexcept;
constexpr Uint128 operator/(Uint128 x, Uint128 y) noexcept;
constexpr Uint128 operator%(Uint128 x, Uint128 y) noexcept;

Arithmetic operators. These all have their normal behaviour for unsigned integers.

constexpr Uint128 Uint128::operator~() const noexcept;
constexpr Uint128& Uint128::operator&=(Uint128 y) noexcept;
constexpr Uint128& Uint128::operator|=(Uint128 y) noexcept;
constexpr Uint128& Uint128::operator^=(Uint128 y) noexcept;
constexpr Uint128& Uint128::operator<<=(int y) noexcept;
constexpr Uint128& Uint128::operator>>=(int y) noexcept;
constexpr Uint128 operator&(Uint128 x, Uint128 y) noexcept;
constexpr Uint128 operator|(Uint128 x, Uint128 y) noexcept;
constexpr Uint128 operator^(Uint128 x, Uint128 y) noexcept;
constexpr Uint128 operator<<(Uint128 x, int y) noexcept;
constexpr Uint128 operator>>(Uint128 x, int y) noexcept;

Bitwise operators. These all have their normal behaviour.

constexpr bool operator==(Uint128 x, Uint128 y) noexcept;
constexpr bool operator!=(Uint128 x, Uint128 y) noexcept;
constexpr bool operator<(Uint128 x, Uint128 y) noexcept;
constexpr bool operator>(Uint128 x, Uint128 y) noexcept;
constexpr bool operator<=(Uint128 x, Uint128 y) noexcept;
constexpr bool operator>=(Uint128 x, Uint128 y) noexcept;
constexpr std::strong_ordering operator<=>(Uint128 x, Uint128 y) noexcept;

Comparison operators. These all have their normal behaviour.

constexpr int Uint128::bit_width() const noexcept;

Returns the number of bits in the value’s binary representation.

constexpr std::pair<Uint128, Uint128>
    Uint128::divide(Uint128 y) const noexcept;

Returns the quotient and remainder of the division. Behaviour is undefined on division by zero.

constexpr std::uint64_t Uint128::high() const noexcept;
constexpr std::uint64_t Uint128::low() const noexcept;

Return the high and low 64-bit halves of the value.

std::string Uint128::str(int base = 10, std::size_t digits = 1) const;
std::string Uint128::hex() const { return str(16, 32); }

Formats the value as a string, in the specified base and with at least the specified number of digits. This will throw std::out_of_range if the base is out of range (2-36). The hex() function displays the full length of the value as hex digits.

template <> struct std::formatter<::RS::Uint128>;

Formatter for Uint128 objects. The only formatting flags recognized are 'b' for binary output, and 'x' for hexadecimal output. The minimum number of digits can also be supplied.

template <> struct std::hash<::RS::Uint128>;

Hash function.