#pragma once

#include <iostream>
#include <iomanip>
#include <string>
#include <vector>
#include <cstring>
#include <cstdint>
#include <bit>
#include <format>

class MD5 {
private:
	static constexpr uint32_t S[64] = {
		7, 12, 17, 22, 7, 12, 17, 22, 7, 12, 17, 22, 7, 12, 17, 22,
		5,  9, 14, 20, 5,  9, 14, 20, 5,  9, 14, 20, 5,  9, 14, 20,
		4, 11, 16, 23, 4, 11, 16, 23, 4, 11, 16, 23, 4, 11, 16, 23,
		6, 10, 15, 21, 6, 10, 15, 21, 6, 10, 15, 21, 6, 10, 15, 21
	};
	
	static constexpr uint32_t K[64] = {
		0xd76aa478, 0xe8c7b756, 0x242070db, 0xc1bdceee,
		0xf57c0faf, 0x4787c62a, 0xa8304613, 0xfd469501,
		0x698098d8, 0x8b44f7af, 0xffff5bb1, 0x895cd7be,
		0x6b901122, 0xfd987193, 0xa679438e, 0x49b40821,
		0xf61e2562, 0xc040b340, 0x265e5a51, 0xe9b6c7aa,
		0xd62f105d, 0x02441453, 0xd8a1e681, 0xe7d3fbc8,
		0x21e1cde6, 0xc33707d6, 0xf4d50d87, 0x455a14ed,
		0xa9e3e905, 0xfcefa3f8, 0x676f02d9, 0x8d2a4c8a,
		0xfffa3942, 0x8771f681, 0x6d9d6122, 0xfde5380c,
		0xa4beea44, 0x4bdecfa9, 0xf6bb4b60, 0xbebfbc70,
		0x289b7ec6, 0xeaa127fa, 0xd4ef3085, 0x04881d05,
		0xd9d4d039, 0xe6db99e5, 0x1fa27cf8, 0xc4ac5665,
		0xf4292244, 0x432aff97, 0xab9423a7, 0xfc93a039,
		0x655b59c3, 0x8f0ccc92, 0xffeff47d, 0x85845dd1,
		0x6fa87e4f, 0xfe2ce6e0, 0xa3014314, 0x4e0811a1,
		0xf7537e82, 0xbd3af235, 0x2ad7d2bb, 0xeb86d391
	};

	uint32_t h[4] = {0x67452301, 0xefcdab89, 0x98badcfe, 0x10325476};

	static uint32_t F(uint32_t x, uint32_t y, uint32_t z) {
		return (x & y) | (~x & z);
	}

	static uint32_t G(uint32_t x, uint32_t y, uint32_t z) {
		return (x & z) | (y & ~z);
	}

	static uint32_t H(uint32_t x, uint32_t y, uint32_t z) {
		return x ^ y ^ z;
	}

	static uint32_t I(uint32_t x, uint32_t y, uint32_t z) {
		return y ^ (x | ~z);
	}

	void processBlock(const uint8_t* block) {
		uint32_t w[16];
		std::memcpy(w, block, 64);

		uint32_t a = h[0], b = h[1], c = h[2], d = h[3];

		for (int i = 0; i < 64; ++i) {
			uint32_t f, g;
			
			if (i < 16) {
				f = F(b, c, d);
				g = i;
			} else if (i < 32) {
				f = G(b, c, d);
				g = (5 * i + 1) % 16;
			} else if (i < 48) {
				f = H(b, c, d);
				g = (3 * i + 5) % 16;
			} else {
				f = I(b, c, d);
				g = (7 * i) % 16;
			}

			f += a + K[i] + w[g];
			a = d;
			d = c;
			c = b;
			b += std::rotl(f, S[i]);
		}

		h[0] += a;
		h[1] += b;
		h[2] += c;
		h[3] += d;
	}

public:
	std::string hash(const std::string& input) {
		h[0] = 0x67452301;
		h[1] = 0xefcdab89;
		h[2] = 0x98badcfe;
		h[3] = 0x10325476;

		std::vector<uint8_t> data(input.begin(), input.end());
		uint64_t originalLength = data.size() * 8;

		data.push_back(0x80);

		while (data.size() % 64 != 56) {
			data.push_back(0);
		}

		for (int i = 0; i < 8; ++i) {
			data.push_back((originalLength >> (i * 8)) & 0xff);
		}

		for (size_t i = 0; i < data.size(); i += 64) {
			processBlock(&data[i]);
		}

		std::string result;
		for (int i = 0; i < 4; ++i) {
			for (int j = 0; j < 4; ++j) {
				uint8_t byte = (h[i] >> (j * 8)) & 0xff;
				result += std::format("{:02x}", byte);
			}
		}

		return result;
	}
};