/*
 * Copyright (C) 2023  Christopher J. Howard
 *
 * This file is part of Antkeeper source code.
 *
 * Antkeeper source code is free software: you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation, either version 3 of the License, or
 * (at your option) any later version.
 *
 * Antkeeper source code is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with Antkeeper source code.  If not, see <http://www.gnu.org/licenses/>.
 */

#ifndef ANTKEEPER_MATH_HASH_PCG_HPP
#define ANTKEEPER_MATH_HASH_PCG_HPP

#include "math/vector.hpp"
#include "math/hash/make-uint.hpp"
#include <cstdint>
#include <type_traits>

namespace math {
namespace hash {

/// @private
template <class T>
constexpr T pcg_multiplier = 0;
template <>
constexpr std::uint8_t pcg_multiplier<std::uint8_t> = 141U;
template <>
constexpr std::uint16_t pcg_multiplier<std::uint16_t> = 12829U;
template <>
constexpr std::uint32_t pcg_multiplier<std::uint32_t> = 747796405UL;
template <>
constexpr std::uint64_t pcg_multiplier<std::uint64_t> = 6364136223846793005ULL;

/// @private
template <class T>
constexpr T pcg_increment = 0;
template <>
constexpr std::uint8_t pcg_increment<std::uint8_t> = 77U;
template <>
constexpr std::uint16_t pcg_increment<std::uint16_t> = 47989U;
template <>
constexpr std::uint32_t pcg_increment<std::uint32_t> = 2891336453UL;
template <>
constexpr std::uint64_t pcg_increment<std::uint64_t> = 1442695040888963407ULL;

/// @private
template <class T>
constexpr T mcg_multiplier = 0;
template <>
constexpr std::uint8_t mcg_multiplier<std::uint8_t> = 217U;
template <>
constexpr std::uint16_t mcg_multiplier<std::uint16_t> = 62169U;
template <>
constexpr std::uint32_t mcg_multiplier<std::uint32_t> = 277803737UL;
template <>
constexpr std::uint64_t mcg_multiplier<std::uint64_t> = 12605985483714917081ULL;

/// @private
template <class T>
[[nodiscard]] constexpr T pcg_uint(T x) noexcept
{
	static_assert(std::is_integral<T>::value && std::is_unsigned<T>::value);
	static_assert(sizeof(T) <= 8);
	
	x = x * pcg_multiplier<T> + pcg_increment<T>;
	x = (x ^ (x >> ((x >> ((sizeof(T) * 8) - (sizeof(T) + 1))) + (sizeof(T) + 1)))) * mcg_multiplier<T>;
	return x ^ (x >> ((sizeof(T) * 16 + 2) / 3));
}

/// @private
template <class T>
[[nodiscard]] inline constexpr vector<T, 1> pcg_uvec1(vector<T, 1> x) noexcept
{
	static_assert(std::is_integral<T>::value && std::is_unsigned<T>::value);
	static_assert(sizeof(T) <= 8);

	x[0] = T(x[0]);
	
	return x;
}

/// @private
template <class T>
[[nodiscard]] constexpr vector<T, 2> pcg_uvec2(vector<T, 2> x) noexcept
{
	static_assert(std::is_integral<T>::value && std::is_unsigned<T>::value);
	static_assert(sizeof(T) <= 8);
	
	x = x * pcg_multiplier<T> + pcg_increment<T>;
	
    x[0] += x[1] * pcg_multiplier<T>;
    x[1] += x[0] * pcg_multiplier<T>;

	x[0] ^= x[0] >> (sizeof(T) * 4);
	x[1] ^= x[1] >> (sizeof(T) * 4);
	
    x[0] += x[1] * pcg_multiplier<T>;
    x[1] += x[0] * pcg_multiplier<T>;
	
	x[0] ^= x[0] >> (sizeof(T) * 4);
	x[1] ^= x[1] >> (sizeof(T) * 4);
	
	return x;
}

/// @private
template <class T>
[[nodiscard]] constexpr vector<T, 3> pcg_uvec3(vector<T, 3> x) noexcept
{
	static_assert(std::is_integral<T>::value && std::is_unsigned<T>::value);
	static_assert(sizeof(T) <= 8);
	
	x = x * pcg_multiplier<T> + pcg_increment<T>;
	
	x[0] += x[1] * x[2];
	x[1] += x[2] * x[0];
	x[2] += x[0] * x[1];
	
	x[0] ^= x[0] >> (sizeof(T) * 4);
	x[1] ^= x[1] >> (sizeof(T) * 4);
	x[2] ^= x[2] >> (sizeof(T) * 4);
	
	x[0] += x[1] * x[2];
	x[1] += x[2] * x[0];
	x[2] += x[0] * x[1];
	
	return x;
}

/// @private
template <class T>
[[nodiscard]] constexpr vector<T, 4> pcg_uvec4(vector<T, 4> x) noexcept
{
	static_assert(std::is_integral<T>::value && std::is_unsigned<T>::value);
	static_assert(sizeof(T) <= 8);
	
	x = x * pcg_multiplier<T> + pcg_increment<T>;
	
	x[0] += x[1] * x[3];
	x[1] += x[2] * x[0];
	x[2] += x[0] * x[1];
	x[3] += x[1] * x[2];
	
	x[0] ^= x[0] >> (sizeof(T) * 4);
	x[1] ^= x[1] >> (sizeof(T) * 4);
	x[2] ^= x[2] >> (sizeof(T) * 4);
	x[3] ^= x[3] >> (sizeof(T) * 4);
	
	x[0] += x[1] * x[3];
	x[1] += x[2] * x[0];
	x[2] += x[0] * x[1];
	x[3] += x[1] * x[2];
	
	return x;
}

/**
 * PCG hash function.
 *
 * @param x Input value.
 *
 * @return Unsigned pseudorandom output value.
 *
 * @warning Floating point and signed input values will be converted to unsigned integers via `static_cast`.
 * @warning Vectors with more than 4 elements are not supported.
 *
 * @see https://en.wikipedia.org/wiki/Permuted_congruential_generator
 * @see O'Neill, M.E. (2014). PCG : A Family of Simple Fast Space-Efficient Statistically Good Algorithms for Random Number Generation.
 * @see Mark Jarzynski and Marc Olano, Hash Functions for GPU Rendering, Journal of Computer Graphics Techniques (JCGT), vol. 9, no. 3, 21-38, 2020.
 */
/// @{
[[nodiscard]] inline constexpr std::uint8_t pcg(std::uint8_t x) noexcept
{
	return pcg_uint<std::uint8_t>(x);
}

[[nodiscard]] inline constexpr std::uint16_t pcg(std::uint16_t x) noexcept
{
	return pcg_uint<std::uint16_t>(x);
}

[[nodiscard]] inline constexpr std::uint32_t pcg(std::uint32_t x) noexcept
{
	return pcg_uint<std::uint32_t>(x);
}

[[nodiscard]] inline constexpr std::uint64_t pcg(std::uint64_t x) noexcept
{
	return pcg_uint<std::uint64_t>(x);
}

[[nodiscard]] inline constexpr std::uint8_t pcg(std::int8_t x) noexcept
{
	return pcg_uint<std::uint8_t>(static_cast<std::uint8_t>(x));
}

[[nodiscard]] inline constexpr std::uint16_t pcg(std::int16_t x) noexcept
{
	return pcg_uint<std::uint16_t>(static_cast<std::uint16_t>(x));
}

[[nodiscard]] inline constexpr std::uint32_t pcg(std::int32_t x) noexcept
{
	return pcg_uint<std::uint32_t>(static_cast<std::uint32_t>(x));
}

[[nodiscard]] inline constexpr std::uint64_t pcg(std::int64_t x) noexcept
{
	return pcg_uint<std::uint64_t>(static_cast<std::uint64_t>(x));
}

[[nodiscard]] inline constexpr std::uint32_t pcg(float x) noexcept
{
	return pcg_uint<std::uint32_t>(static_cast<std::uint32_t>(x));
}

[[nodiscard]] inline constexpr std::uint64_t pcg(double x) noexcept
{
	return pcg_uint<std::uint64_t>(static_cast<std::uint64_t>(x));
}

template <class T, std::size_t N>
[[nodiscard]] inline constexpr vector<make_uint_t<T>, N> pcg(const vector<T, N>& x) noexcept
{
	static_assert(N > 0 && N < 5, "PCG hash only supports vectors with 1-4 elements.");
	
	if constexpr (N == 1)
		return pcg_uvec1<make_uint_t<T>>(vector<make_uint_t<T>, N>(x));
	else if constexpr (N == 2)
		return pcg_uvec2<make_uint_t<T>>(vector<make_uint_t<T>, N>(x));
	else if constexpr (N == 3)
		return pcg_uvec3<make_uint_t<T>>(vector<make_uint_t<T>, N>(x));
	else
		return pcg_uvec4<make_uint_t<T>>(vector<make_uint_t<T>, N>(x));
}
/// @}

} // namespace hash
} // namespace math

#endif // ANTKEEPER_MATH_HASH_PCG_HPP