_worker.js

// @ts-ignore
import { connect } from 'cloudflare:sockets';

// How to generate your own UUID:
// [Windows] Press "Win + R", input cmd and run:  Powershell -NoExit -Command "[guid]::NewGuid()"
let userID = 'd342d11e-d424-4583-b36e-524ab1f0afa4';

const proxyIPs = ['23.162.136.169', 'cdn.xn--b6gac.eu.org', 'cdn-all.xn--b6gac.eu.org', 'edgetunnel.anycast.eu.org'];

let proxyIP = proxyIPs[Math.floor(Math.random() * proxyIPs.length)];

let dohURL = 'https://sky.rethinkdns.com/1:-Pf_____9_8A_AMAIgE8kMABVDDmKOHTAKg='; // https://cloudflare-dns.com/dns-query or https://dns.google/dns-query

// v2board api environment variables (optional) deprecated, please use planetscale.com instead

if (!isValidUUID(userID)) {
	throw new Error('uuid is invalid');
}

export default {
	/**
	 * @param {import("@cloudflare/workers-types").Request} request
	 * @param {{UUID: string, PROXYIP: string, DNS_RESOLVER_URL: string, NODE_ID: int, API_HOST: string, API_TOKEN: string}} env
	 * @param {import("@cloudflare/workers-types").ExecutionContext} ctx
	 * @returns {Promise<Response>}
	 */
	async fetch(request, env, ctx) {
		// uuid_validator(request);
		try {
			userID = env.UUID || userID;
			proxyIP = env.PROXYIP || proxyIP;
			dohURL = env.DNS_RESOLVER_URL || dohURL;
			let userID_Path = userID;
			if (userID.includes(',')) {
				userID_Path = userID.split(',')[0];
			}
			const upgradeHeader = request.headers.get('Upgrade');
			if (!upgradeHeader || upgradeHeader !== 'websocket') {
				const url = new URL(request.url);
				switch (url.pathname) {
					case `/cf`: {
						return new Response(JSON.stringify(request.cf, null, 4), {
							status: 200,
							headers: {
								"Content-Type": "application/json;charset=utf-8",
							},
						});
					}
					case `/${userID_Path}`: {
						const vlessConfig = getVLESSConfig(userID, request.headers.get('Host'));
						return new Response(`${vlessConfig}`, {
							status: 200,
							headers: {
								"Content-Type": "text/html; charset=utf-8",
							}
						});
					};
					case `/sub/${userID_Path}`: {
						const url = new URL(request.url);
						const searchParams = url.searchParams;
						const vlessSubConfig = createVLESSSub(userID, request.headers.get('Host'));
						// Construct and return response object
						return new Response(btoa(vlessSubConfig), {
							status: 200,
							headers: {
								"Content-Type": "text/plain;charset=utf-8",
							}
						});
					};
					default:
						// return new Response('Not found', { status: 404 });
						// For any other path, reverse proxy to 'ramdom website' and return the original response, caching it in the process
						const randomHostname = cn_hostnames[Math.floor(Math.random() * cn_hostnames.length)];
						const newHeaders = new Headers(request.headers);
						newHeaders.set('cf-connecting-ip', '1.2.3.4');
						newHeaders.set('x-forwarded-for', '1.2.3.4');
						newHeaders.set('x-real-ip', '1.2.3.4');
						newHeaders.set('referer', 'https://www.google.com/search?q=edtunnel');
						// Use fetch to proxy the request to 15 different domains
						const proxyUrl = 'https://' + randomHostname + url.pathname + url.search;
						let modifiedRequest = new Request(proxyUrl, {
							method: request.method,
							headers: newHeaders,
							body: request.body,
							redirect: 'manual',
						});
						const proxyResponse = await fetch(modifiedRequest, { redirect: 'manual' });
						// Check for 302 or 301 redirect status and return an error response
						if ([301, 302].includes(proxyResponse.status)) {
							return new Response(`Redirects to ${randomHostname} are not allowed.`, {
								status: 403,
								statusText: 'Forbidden',
							});
						}
						// Return the response from the proxy server
						return proxyResponse;
				}
			} else {
				return await vlessOverWSHandler(request);
			}
		} catch (err) {
			/** @type {Error} */ let e = err;
			return new Response(e.toString());
		}
	},
};

export async function uuid_validator(request) {
	const hostname = request.headers.get('Host');
	const currentDate = new Date();

	const subdomain = hostname.split('.')[0];
	const year = currentDate.getFullYear();
	const month = String(currentDate.getMonth() + 1).padStart(2, '0');
	const day = String(currentDate.getDate()).padStart(2, '0');

	const formattedDate = `${year}-${month}-${day}`;

	// const daliy_sub = formattedDate + subdomain
	const hashHex = await hashHex_f(subdomain);
	// subdomain string contains timestamps utc and uuid string TODO.
	console.log(hashHex, subdomain, formattedDate);
}

export async function hashHex_f(string) {
	const encoder = new TextEncoder();
	const data = encoder.encode(string);
	const hashBuffer = await crypto.subtle.digest('SHA-256', data);
	const hashArray = Array.from(new Uint8Array(hashBuffer));
	const hashHex = hashArray.map(byte => byte.toString(16).padStart(2, '0')).join('');
	return hashHex;
}

/**
 * Handles VLESS over WebSocket requests by creating a WebSocket pair, accepting the WebSocket connection, and processing the VLESS header.
 * @param {import("@cloudflare/workers-types").Request} request The incoming request object.
 * @returns {Promise<Response>} A Promise that resolves to a WebSocket response object.
 */
async function vlessOverWSHandler(request) {
	const webSocketPair = new WebSocketPair();
	const [client, webSocket] = Object.values(webSocketPair);
	webSocket.accept();

	let address = '';
	let portWithRandomLog = '';
	let currentDate = new Date();
	const log = (/** @type {string} */ info, /** @type {string | undefined} */ event) => {
		console.log(`[${currentDate} ${address}:${portWithRandomLog}] ${info}`, event || '');
	};
	const earlyDataHeader = request.headers.get('sec-websocket-protocol') || '';

	const readableWebSocketStream = makeReadableWebSocketStream(webSocket, earlyDataHeader, log);

	/** @type {{ value: import("@cloudflare/workers-types").Socket | null}}*/
	let remoteSocketWapper = {
		value: null,
	};
	let udpStreamWrite = null;
	let isDns = false;

	// ws --> remote
	readableWebSocketStream.pipeTo(new WritableStream({
		async write(chunk, controller) {
			if (isDns && udpStreamWrite) {
				return udpStreamWrite(chunk);
			}
			if (remoteSocketWapper.value) {
				const writer = remoteSocketWapper.value.writable.getWriter()
				await writer.write(chunk);
				writer.releaseLock();
				return;
			}

			const {
				hasError,
				message,
				portRemote = 443,
				addressRemote = '',
				rawDataIndex,
				vlessVersion = new Uint8Array([0, 0]),
				isUDP,
			} = processVlessHeader(chunk, userID);
			address = addressRemote;
			portWithRandomLog = `${portRemote} ${isUDP ? 'udp' : 'tcp'} `;
			if (hasError) {
				// controller.error(message);
				throw new Error(message); // cf seems has bug, controller.error will not end stream
				// webSocket.close(1000, message);
				return;
			}

			// If UDP and not DNS port, close it
			if (isUDP && portRemote !== 53) {
				throw new Error('UDP proxy only enabled for DNS which is port 53');
				// cf seems has bug, controller.error will not end stream
			}

			if (isUDP && portRemote === 53) {
				isDns = true;
			}

			// ["version", "附加信息长度 N"]
			const vlessResponseHeader = new Uint8Array([vlessVersion[0], 0]);
			const rawClientData = chunk.slice(rawDataIndex);

			// TODO: support udp here when cf runtime has udp support
			if (isDns) {
				const { write } = await handleUDPOutBound(webSocket, vlessResponseHeader, log);
				udpStreamWrite = write;
				udpStreamWrite(rawClientData);
				return;
			}
			handleTCPOutBound(remoteSocketWapper, addressRemote, portRemote, rawClientData, webSocket, vlessResponseHeader, log);
		},
		close() {
			log(`readableWebSocketStream is close`);
		},
		abort(reason) {
			log(`readableWebSocketStream is abort`, JSON.stringify(reason));
		},
	})).catch((err) => {
		log('readableWebSocketStream pipeTo error', err);
	});

	return new Response(null, {
		status: 101,
		webSocket: client,
	});
}

/**
 * Handles outbound TCP connections.
 *
 * @param {any} remoteSocket 
 * @param {string} addressRemote The remote address to connect to.
 * @param {number} portRemote The remote port to connect to.
 * @param {Uint8Array} rawClientData The raw client data to write.
 * @param {import("@cloudflare/workers-types").WebSocket} webSocket The WebSocket to pass the remote socket to.
 * @param {Uint8Array} vlessResponseHeader The VLESS response header.
 * @param {function} log The logging function.
 * @returns {Promise<void>} The remote socket.
 */
async function handleTCPOutBound(remoteSocket, addressRemote, portRemote, rawClientData, webSocket, vlessResponseHeader, log,) {

	/**
	 * Connects to a given address and port and writes data to the socket.
	 * @param {string} address The address to connect to.
	 * @param {number} port The port to connect to.
	 * @returns {Promise<import("@cloudflare/workers-types").Socket>} A Promise that resolves to the connected socket.
	 */
	async function connectAndWrite(address, port) {
		/** @type {import("@cloudflare/workers-types").Socket} */
		const tcpSocket = connect({
			hostname: address,
			port: port,
		});
		remoteSocket.value = tcpSocket;
		log(`connected to ${address}:${port}`);
		const writer = tcpSocket.writable.getWriter();
		await writer.write(rawClientData); // first write, nomal is tls client hello
		writer.releaseLock();
		return tcpSocket;
	}

	/**
	 * Retries connecting to the remote address and port if the Cloudflare socket has no incoming data.
	 * @returns {Promise<void>} A Promise that resolves when the retry is complete.
	 */
	async function retry() {
		const tcpSocket = await connectAndWrite(proxyIP || addressRemote, portRemote)
		tcpSocket.closed.catch(error => {
			console.log('retry tcpSocket closed error', error);
		}).finally(() => {
			safeCloseWebSocket(webSocket);
		})
		remoteSocketToWS(tcpSocket, webSocket, vlessResponseHeader, null, log);
	}

	const tcpSocket = await connectAndWrite(addressRemote, portRemote);

	// when remoteSocket is ready, pass to websocket
	// remote--> ws
	remoteSocketToWS(tcpSocket, webSocket, vlessResponseHeader, retry, log);
}

/**
 * Creates a readable stream from a WebSocket server, allowing for data to be read from the WebSocket.
 * @param {import("@cloudflare/workers-types").WebSocket} webSocketServer The WebSocket server to create the readable stream from.
 * @param {string} earlyDataHeader The header containing early data for WebSocket 0-RTT.
 * @param {(info: string)=> void} log The logging function.
 * @returns {ReadableStream} A readable stream that can be used to read data from the WebSocket.
 */
function makeReadableWebSocketStream(webSocketServer, earlyDataHeader, log) {
	let readableStreamCancel = false;
	const stream = new ReadableStream({
		start(controller) {
			webSocketServer.addEventListener('message', (event) => {
				const message = event.data;
				controller.enqueue(message);
			});

			webSocketServer.addEventListener('close', () => {
				safeCloseWebSocket(webSocketServer);
				controller.close();
			});

			webSocketServer.addEventListener('error', (err) => {
				log('webSocketServer has error');
				controller.error(err);
			});
			const { earlyData, error } = base64ToArrayBuffer(earlyDataHeader);
			if (error) {
				controller.error(error);
			} else if (earlyData) {
				controller.enqueue(earlyData);
			}
		},

		pull(controller) {
			// if ws can stop read if stream is full, we can implement backpressure
			// https://streams.spec.whatwg.org/#example-rs-push-backpressure
		},

		cancel(reason) {
			log(`ReadableStream was canceled, due to ${reason}`)
			readableStreamCancel = true;
			safeCloseWebSocket(webSocketServer);
		}
	});

	return stream;
}

// https://xtls.github.io/development/protocols/vless.html
// https://github.com/zizifn/excalidraw-backup/blob/main/v2ray-protocol.excalidraw

/**
 * Processes the VLESS header buffer and returns an object with the relevant information.
 * @param {ArrayBuffer} vlessBuffer The VLESS header buffer to process.
 * @param {string} userID The user ID to validate against the UUID in the VLESS header.
 * @returns {{
 *  hasError: boolean,
 *  message?: string,
 *  addressRemote?: string,
 *  addressType?: number,
 *  portRemote?: number,
 *  rawDataIndex?: number,
 *  vlessVersion?: Uint8Array,
 *  isUDP?: boolean
 * }} An object with the relevant information extracted from the VLESS header buffer.
 */
function processVlessHeader(vlessBuffer, userID) {
	if (vlessBuffer.byteLength < 24) {
		return {
			hasError: true,
			message: 'invalid data',
		};
	}

	const version = new Uint8Array(vlessBuffer.slice(0, 1));
	let isValidUser = false;
	let isUDP = false;
	const slicedBuffer = new Uint8Array(vlessBuffer.slice(1, 17));
	const slicedBufferString = stringify(slicedBuffer);
	// check if userID is valid uuid or uuids split by , and contains userID in it otherwise return error message to console
	const uuids = userID.includes(',') ? userID.split(",") : [userID];
	// uuid_validator(hostName, slicedBufferString);


	// isValidUser = uuids.some(userUuid => slicedBufferString === userUuid.trim());
	isValidUser = uuids.some(userUuid => slicedBufferString === userUuid.trim()) || uuids.length === 1 && slicedBufferString === uuids[0].trim();

	console.log(`userID: ${slicedBufferString}`);

	if (!isValidUser) {
		return {
			hasError: true,
			message: 'invalid user',
		};
	}

	const optLength = new Uint8Array(vlessBuffer.slice(17, 18))[0];
	//skip opt for now

	const command = new Uint8Array(
		vlessBuffer.slice(18 + optLength, 18 + optLength + 1)
	)[0];

	// 0x01 TCP
	// 0x02 UDP
	// 0x03 MUX
	if (command === 1) {
		isUDP = false;
	} else if (command === 2) {
		isUDP = true;
	} else {
		return {
			hasError: true,
			message: `command ${command} is not support, command 01-tcp,02-udp,03-mux`,
		};
	}
	const portIndex = 18 + optLength + 1;
	const portBuffer = vlessBuffer.slice(portIndex, portIndex + 2);
	// port is big-Endian in raw data etc 80 == 0x005d
	const portRemote = new DataView(portBuffer).getUint16(0);

	let addressIndex = portIndex + 2;
	const addressBuffer = new Uint8Array(
		vlessBuffer.slice(addressIndex, addressIndex + 1)
	);

	// 1--> ipv4  addressLength =4
	// 2--> domain name addressLength=addressBuffer[1]
	// 3--> ipv6  addressLength =16
	const addressType = addressBuffer[0];
	let addressLength = 0;
	let addressValueIndex = addressIndex + 1;
	let addressValue = '';
	switch (addressType) {
		case 1:
			addressLength = 4;
			addressValue = new Uint8Array(
				vlessBuffer.slice(addressValueIndex, addressValueIndex + addressLength)
			).join('.');
			break;
		case 2:
			addressLength = new Uint8Array(
				vlessBuffer.slice(addressValueIndex, addressValueIndex + 1)
			)[0];
			addressValueIndex += 1;
			addressValue = new TextDecoder().decode(
				vlessBuffer.slice(addressValueIndex, addressValueIndex + addressLength)
			);
			break;
		case 3:
			addressLength = 16;
			const dataView = new DataView(
				vlessBuffer.slice(addressValueIndex, addressValueIndex + addressLength)
			);
			// 2001:0db8:85a3:0000:0000:8a2e:0370:7334
			const ipv6 = [];
			for (let i = 0; i < 8; i++) {
				ipv6.push(dataView.getUint16(i * 2).toString(16));
			}
			addressValue = ipv6.join(':');
			// seems no need add [] for ipv6
			break;
		default:
			return {
				hasError: true,
				message: `invild  addressType is ${addressType}`,
			};
	}
	if (!addressValue) {
		return {
			hasError: true,
			message: `addressValue is empty, addressType is ${addressType}`,
		};
	}

	return {
		hasError: false,
		addressRemote: addressValue,
		addressType,
		portRemote,
		rawDataIndex: addressValueIndex + addressLength,
		vlessVersion: version,
		isUDP,
	};
}


/**
 * Converts a remote socket to a WebSocket connection.
 * @param {import("@cloudflare/workers-types").Socket} remoteSocket The remote socket to convert.
 * @param {import("@cloudflare/workers-types").WebSocket} webSocket The WebSocket to connect to.
 * @param {ArrayBuffer | null} vlessResponseHeader The VLESS response header.
 * @param {(() => Promise<void>) | null} retry The function to retry the connection if it fails.
 * @param {(info: string) => void} log The logging function.
 * @returns {Promise<void>} A Promise that resolves when the conversion is complete.
 */
async function remoteSocketToWS(remoteSocket, webSocket, vlessResponseHeader, retry, log) {
	// remote--> ws
	let remoteChunkCount = 0;
	let chunks = [];
	/** @type {ArrayBuffer | null} */
	let vlessHeader = vlessResponseHeader;
	let hasIncomingData = false; // check if remoteSocket has incoming data
	await remoteSocket.readable
		.pipeTo(
			new WritableStream({
				start() {
				},
				/**
				 * 
				 * @param {Uint8Array} chunk 
				 * @param {*} controller 
				 */
				async write(chunk, controller) {
					hasIncomingData = true;
					remoteChunkCount++;
					if (webSocket.readyState !== WS_READY_STATE_OPEN) {
						controller.error(
							'webSocket.readyState is not open, maybe close'
						);
					}
					if (vlessHeader) {
						webSocket.send(await new Blob([vlessHeader, chunk]).arrayBuffer());
						vlessHeader = null;
					} else {
						// console.log(`remoteSocketToWS send chunk ${chunk.byteLength}`);
						// seems no need rate limit this, CF seems fix this??..
						// if (remoteChunkCount > 20000) {
						// 	// cf one package is 4096 byte(4kb),  4096 * 20000 = 80M
						// 	await delay(1);
						// }
						webSocket.send(chunk);
					}
				},
				close() {
					log(`remoteConnection!.readable is close with hasIncomingData is ${hasIncomingData}`);
					// safeCloseWebSocket(webSocket); // no need server close websocket frist for some case will casue HTTP ERR_CONTENT_LENGTH_MISMATCH issue, client will send close event anyway.
				},
				abort(reason) {
					console.error(`remoteConnection!.readable abort`, reason);
				},
			})
		)
		.catch((error) => {
			console.error(
				`remoteSocketToWS has exception `,
				error.stack || error
			);
			safeCloseWebSocket(webSocket);
		});

	// seems is cf connect socket have error,
	// 1. Socket.closed will have error
	// 2. Socket.readable will be close without any data coming
	if (hasIncomingData === false && retry) {
		log(`retry`)
		retry();
	}
}

/**
 * Decodes a base64 string into an ArrayBuffer.
 * @param {string} base64Str The base64 string to decode.
 * @returns {{earlyData: ArrayBuffer|null, error: Error|null}} An object containing the decoded ArrayBuffer or null if there was an error, and any error that occurred during decoding or null if there was no error.
 */
function base64ToArrayBuffer(base64Str) {
	if (!base64Str) {
		return { earlyData: null, error: null };
	}
	try {
		// go use modified Base64 for URL rfc4648 which js atob not support
		base64Str = base64Str.replace(/-/g, '+').replace(/_/g, '/');
		const decode = atob(base64Str);
		const arryBuffer = Uint8Array.from(decode, (c) => c.charCodeAt(0));
		return { earlyData: arryBuffer.buffer, error: null };
	} catch (error) {
		return { earlyData: null, error };
	}
}

/**
 * Checks if a given string is a valid UUID.
 * Note: This is not a real UUID validation.
 * @param {string} uuid The string to validate as a UUID.
 * @returns {boolean} True if the string is a valid UUID, false otherwise.
 */
function isValidUUID(uuid) {
	const uuidRegex = /^[0-9a-f]{8}-[0-9a-f]{4}-[4][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$/i;
	return uuidRegex.test(uuid);
}

const WS_READY_STATE_OPEN = 1;
const WS_READY_STATE_CLOSING = 2;
/**
 * Closes a WebSocket connection safely without throwing exceptions.
 * @param {import("@cloudflare/workers-types").WebSocket} socket The WebSocket connection to close.
 */
function safeCloseWebSocket(socket) {
	try {
		if (socket.readyState === WS_READY_STATE_OPEN || socket.readyState === WS_READY_STATE_CLOSING) {
			socket.close();
		}
	} catch (error) {
		console.error('safeCloseWebSocket error', error);
	}
}

const byteToHex = [];

for (let i = 0; i < 256; ++i) {
	byteToHex.push((i + 256).toString(16).slice(1));
}

function unsafeStringify(arr, offset = 0) {
	return (byteToHex[arr[offset + 0]] + byteToHex[arr[offset + 1]] + byteToHex[arr[offset + 2]] + byteToHex[arr[offset + 3]] + "-" + byteToHex[arr[offset + 4]] + byteToHex[arr[offset + 5]] + "-" + byteToHex[arr[offset + 6]] + byteToHex[arr[offset + 7]] + "-" + byteToHex[arr[offset + 8]] + byteToHex[arr[offset + 9]] + "-" + byteToHex[arr[offset + 10]] + byteToHex[arr[offset + 11]] + byteToHex[arr[offset + 12]] + byteToHex[arr[offset + 13]] + byteToHex[arr[offset + 14]] + byteToHex[arr[offset + 15]]).toLowerCase();
}

function stringify(arr, offset = 0) {
	const uuid = unsafeStringify(arr, offset);
	if (!isValidUUID(uuid)) {
		throw TypeError("Stringified UUID is invalid");
	}
	return uuid;
}


/**
 * Handles outbound UDP traffic by transforming the data into DNS queries and sending them over a WebSocket connection.
 * @param {import("@cloudflare/workers-types").WebSocket} webSocket The WebSocket connection to send the DNS queries over.
 * @param {ArrayBuffer} vlessResponseHeader The VLESS response header.
 * @param {(string) => void} log The logging function.
 * @returns {{write: (chunk: Uint8Array) => void}} An object with a write method that accepts a Uint8Array chunk to write to the transform stream.
 */
async function handleUDPOutBound(webSocket, vlessResponseHeader, log) {

	let isVlessHeaderSent = false;
	const transformStream = new TransformStream({
		start(controller) {

		},
		transform(chunk, controller) {
			// udp message 2 byte is the the length of udp data
			// TODO: this should have bug, beacsue maybe udp chunk can be in two websocket message
			for (let index = 0; index < chunk.byteLength;) {
				const lengthBuffer = chunk.slice(index, index + 2);
				const udpPakcetLength = new DataView(lengthBuffer).getUint16(0);
				const udpData = new Uint8Array(
					chunk.slice(index + 2, index + 2 + udpPakcetLength)
				);
				index = index + 2 + udpPakcetLength;
				controller.enqueue(udpData);
			}
		},
		flush(controller) {
		}
	});

	// only handle dns udp for now
	transformStream.readable.pipeTo(new WritableStream({
		async write(chunk) {
			const resp = await fetch(dohURL, // dns server url
				{
					method: 'POST',
					headers: {
						'content-type': 'application/dns-message',
					},
					body: chunk,
				})
			const dnsQueryResult = await resp.arrayBuffer();
			const udpSize = dnsQueryResult.byteLength;
			// console.log([...new Uint8Array(dnsQueryResult)].map((x) => x.toString(16)));
			const udpSizeBuffer = new Uint8Array([(udpSize >> 8) & 0xff, udpSize & 0xff]);
			if (webSocket.readyState === WS_READY_STATE_OPEN) {
				log(`doh success and dns message length is ${udpSize}`);
				if (isVlessHeaderSent) {
					webSocket.send(await new Blob([udpSizeBuffer, dnsQueryResult]).arrayBuffer());
				} else {
					webSocket.send(await new Blob([vlessResponseHeader, udpSizeBuffer, dnsQueryResult]).arrayBuffer());
					isVlessHeaderSent = true;
				}
			}
		}
	})).catch((error) => {
		log('dns udp has error' + error)
	});

	const writer = transformStream.writable.getWriter();

	return {
		/**
		 * 
		 * @param {Uint8Array} chunk 
		 */
		write(chunk) {
			writer.write(chunk);
		}
	};
}

/**
 *
 * @param {string} userID - single or comma separated userIDs
 * @param {string | null} hostName
 * @returns {string}
 */
function getVLESSConfig(userIDs, hostName) {
	const commonUrlPart = `:443?encryption=none&security=tls&sni=${hostName}&fp=randomized&type=ws&host=${hostName}&path=%2F%3Fed%3D2048#${hostName}`;
	const separator = "---------------------------------------------------------------";
	const hashSeparator = "################################################################";

	// Split the userIDs into an array
	let userIDArray = userIDs.split(',');

	// Prepare output array
	let output = [];
	let header = [];
	const sublink = `https://${hostName}/sub/${userIDArray[0]}?format=clash`
	const clash_link = `https://api.v1.mk/sub?target=clash&url=${encodeURIComponent(sublink)}&insert=false&emoji=true&list=false&tfo=false&scv=true&fdn=false&sort=false&new_name=true`;
	header.push(`\n<p align="center"><img src="https://cloudflare-ipfs.com/ipfs/bafybeigd6i5aavwpr6wvnwuyayklq3omonggta4x2q7kpmgafj357nkcky" alt="图片描述" style="margin-bottom: -50px;">`);
	header.push(`\n<b style=" font-size: 15px;" >Welcome! This function generates configuration for VLESS protocol. If you found this useful, please check our GitHub project for more:</b>\n`);
	header.push(`<b style=" font-size: 15px;" >欢迎！这是生成 VLESS 协议的配置。如果您发现这个项目很好用，请查看我们的 GitHub 项目给我一个star：</b>\n`);
	header.push(`\n<a href="https://github.com/3Kmfi6HP/EDtunnel" target="_blank">EDtunnel - https://github.com/3Kmfi6HP/EDtunnel</a>\n`);
	header.push(`\n<iframe src="https://ghbtns.com/github-btn.html?user=USERNAME&repo=REPOSITORY&type=star&count=true&size=large" frameborder="0" scrolling="0" width="170" height="30" title="GitHub"></iframe>\n\n`.replace(/USERNAME/g, "3Kmfi6HP").replace(/REPOSITORY/g, "EDtunnel"));
	header.push(`<a href="//${hostName}/sub/${userIDArray[0]}" target="_blank">VLESS 节点订阅连接</a>\n<a href="clash://install-config?url=${encodeURIComponent(clash_link)}" target="_blank">Clash for Windows 节点订阅连接</a>\n<a href="${clash_link}" target="_blank">Clash 节点订阅连接</a>\n<a href="https://sub.xf.free.hr/auto?host=${hostName}&uuid=${userIDArray[0]}" target="_blank">优选IP自动节点订阅</a></p>\n`);
	header.push(``);

	// Generate output string for each userID
	userIDArray.forEach((userID) => {
		const vlessMain = `vless://${userID}@${hostName}${commonUrlPart}`;
		const vlessSec = `vless://${userID}@${proxyIP}${commonUrlPart}`;
		output.push(`UUID: ${userID}`);
		output.push(`${hashSeparator}\nv2ray default ip\n${separator}\n${vlessMain}\n${separator}`);
		output.push(`${hashSeparator}\nv2ray with best ip\n${separator}\n${vlessSec}\n${separator}`);
	});
	output.push(`${hashSeparator}\n# Clash Proxy Provider 配置格式(configuration format)\nproxy-groups:\n  - name: UseProvider\n	type: select\n	use:\n	  - provider1\n	proxies:\n	  - Proxy\n	  - DIRECT\nproxy-providers:\n  provider1:\n	type: http\n	url: https://${hostName}/sub/${userIDArray[0]}?format=clash\n	interval: 3600\n	path: ./provider1.yaml\n	health-check:\n	  enable: true\n	  interval: 600\n	  # lazy: true\n	  url: http://www.gstatic.com/generate_204\n\n${hashSeparator}`);

	// HTML Head with CSS
	const htmlHead = `
    <head>
        <title>EDtunnel: VLESS configuration</title>
        <meta name="description" content="This is a tool for generating VLESS protocol configurations. Give us a star on GitHub https://github.com/3Kmfi6HP/EDtunnel if you found it useful!">
		<meta name="keywords" content="EDtunnel, cloudflare pages, cloudflare worker, severless">
        <meta name="viewport" content="width=device-width, initial-scale=1">
		<meta property="og:site_name" content="EDtunnel: VLESS configuration" />
        <meta property="og:type" content="website" />
        <meta property="og:title" content="EDtunnel - VLESS configuration and subscribe output" />
        <meta property="og:description" content="Use cloudflare pages and worker severless to implement vless protocol" />
        <meta property="og:url" content="https://${hostName}/" />
        <meta property="og:image" content="https://api.qrserver.com/v1/create-qr-code/?size=500x500&data=${encodeURIComponent(`vless://${userIDs.split(',')[0]}@${hostName}${commonUrlPart}`)}" />
        <meta name="twitter:card" content="summary_large_image" />
        <meta name="twitter:title" content="EDtunnel - VLESS configuration and subscribe output" />
        <meta name="twitter:description" content="Use cloudflare pages and worker severless to implement vless protocol" />
        <meta name="twitter:url" content="https://${hostName}/" />
        <meta name="twitter:image" content="https://cloudflare-ipfs.com/ipfs/bafybeigd6i5aavwpr6wvnwuyayklq3omonggta4x2q7kpmgafj357nkcky" />
        <meta property="og:image:width" content="1500" />
        <meta property="og:image:height" content="1500" />

        <style>
        body {
            font-family: Arial, sans-serif;
            background-color: #f0f0f0;
            color: #333;
            padding: 10px;
        }

        a {
            color: #1a0dab;
            text-decoration: none;
        }
		img {
			max-width: 100%;
			height: auto;
		}
		
        pre {
            white-space: pre-wrap;
            word-wrap: break-word;
            background-color: #fff;
            border: 1px solid #ddd;
            padding: 15px;
            margin: 10px 0;
        }
		/* Dark mode */
        @media (prefers-color-scheme: dark) {
            body {
                background-color: #333;
                color: #f0f0f0;
            }

            a {
                color: #9db4ff;
            }

            pre {
                background-color: #282a36;
                border-color: #6272a4;
            }
        }
        </style>
    </head>
    `;

	// Join output with newlines, wrap inside <html> and <body>
	return `
    <html>
    ${htmlHead}
    <body>
    <pre style="
    background-color: transparent;
    border: none;
">${header.join('')}</pre><pre>${output.join('\n')}</pre>
    </body>
</html>`;
}


function createVLESSSub(userID_Path, hostName) {
	let portArray_http = [80, 8080, 8880, 2052, 2086, 2095, 2082];
	let portArray_https = [443, 8443, 2053, 2096, 2087, 2083];

	// Split the userIDs into an array
	let userIDArray = userID_Path.includes(',') ? userID_Path.split(',') : [userID_Path];

	// Prepare output array
	let output = [];

	// Generate output string for each userID
	userIDArray.forEach((userID) => {
		// Check if the hostName is a Cloudflare Pages domain, if not, generate HTTP configurations
		// reasons: pages.dev not support http only https
		if (!hostName.includes('pages.dev')) {
			// Iterate over all ports for http
			portArray_http.forEach((port) => {
				const commonUrlPart_http = `:${port}?encryption=none&security=none&fp=random&type=ws&host=${hostName}&path=%2F%3Fed%3D2048#${hostName}-HTTP-${port}`;
				const vlessMainHttp = `vless://${userID}@${hostName}${commonUrlPart_http}`;

				// For each proxy IP, generate a VLESS configuration and add to output
				proxyIPs.forEach((proxyIP) => {
					const vlessSecHttp = `vless://${userID}@${proxyIP}${commonUrlPart_http}-${proxyIP}-EDtunnel`;
					output.push(`${vlessMainHttp}`);
					output.push(`${vlessSecHttp}`);
				});
			});
		}
		// Iterate over all ports for https
		portArray_https.forEach((port) => {
			const commonUrlPart_https = `:${port}?encryption=none&security=tls&sni=${hostName}&fp=random&type=ws&host=${hostName}&path=%2F%3Fed%3D2048#${hostName}-HTTPS-${port}`;
			const vlessMainHttps = `vless://${userID}@${hostName}${commonUrlPart_https}`;

			// For each proxy IP, generate a VLESS configuration and add to output
			proxyIPs.forEach((proxyIP) => {
				const vlessSecHttps = `vless://${userID}@${proxyIP}${commonUrlPart_https}-${proxyIP}-EDtunnel`;
				output.push(`${vlessMainHttps}`);
				output.push(`${vlessSecHttps}`);
			});
		});
	});

	// Join output with newlines
	return output.join('\n');
}

const cn_hostnames = [
	'weibo.com',                // Weibo - A popular social media platform
	'www.baidu.com',            // Baidu - The largest search engine in China
	'www.qq.com',               // QQ - A widely used instant messaging platform
	'www.taobao.com',           // Taobao - An e-commerce website owned by Alibaba Group
	'www.jd.com',               // JD.com - One of the largest online retailers in China
	'www.sina.com.cn',          // Sina - A Chinese online media company
	'www.sohu.com',             // Sohu - A Chinese internet service provider
	'www.tmall.com',            // Tmall - An online retail platform owned by Alibaba Group
	'www.163.com',              // NetEase Mail - One of the major email providers in China
	'www.zhihu.com',            // Zhihu - A popular question-and-answer website
	'www.youku.com',            // Youku - A Chinese video sharing platform
	'www.xinhuanet.com',        // Xinhua News Agency - Official news agency of China
	'www.douban.com',           // Douban - A Chinese social networking service
	'www.meituan.com',          // Meituan - A Chinese group buying website for local services
	'www.toutiao.com',          // Toutiao - A news and information content platform
	'www.ifeng.com',            // iFeng - A popular news website in China
	'www.autohome.com.cn',      // Autohome - A leading Chinese automobile online platform
	'www.360.cn',               // 360 - A Chinese internet security company
	'www.douyin.com',           // Douyin - A Chinese short video platform
	'www.kuaidi100.com',        // Kuaidi100 - A Chinese express delivery tracking service
	'www.wechat.com',           // WeChat - A popular messaging and social media app
	'www.csdn.net',             // CSDN - A Chinese technology community website
	'www.imgo.tv',              // ImgoTV - A Chinese live streaming platform
	'www.aliyun.com',           // Alibaba Cloud - A Chinese cloud computing company
	'www.eyny.com',             // Eyny - A Chinese multimedia resource-sharing website
	'www.mgtv.com',             // MGTV - A Chinese online video platform
	'www.xunlei.com',           // Xunlei - A Chinese download manager and torrent client
	'www.hao123.com',           // Hao123 - A Chinese web directory service
	'www.bilibili.com',         // Bilibili - A Chinese video sharing and streaming platform
	'www.youth.cn',             // Youth.cn - A China Youth Daily news portal
	'www.hupu.com',             // Hupu - A Chinese sports community and forum
	'www.youzu.com',            // Youzu Interactive - A Chinese game developer and publisher
	'www.panda.tv',             // Panda TV - A Chinese live streaming platform
	'www.tudou.com',            // Tudou - A Chinese video-sharing website
	'www.zol.com.cn',           // ZOL - A Chinese electronics and gadgets website
	'www.toutiao.io',           // Toutiao - A news and information app
	'www.tiktok.com',           // TikTok - A Chinese short-form video app
	'www.netease.com',          // NetEase - A Chinese internet technology company
	'www.cnki.net',             // CNKI - China National Knowledge Infrastructure, an information aggregator
	'www.zhibo8.cc',            // Zhibo8 - A website providing live sports streams
	'www.zhangzishi.cc',        // Zhangzishi - Personal website of Zhang Zishi, a public intellectual in China
	'www.xueqiu.com',           // Xueqiu - A Chinese online social platform for investors and traders
	'www.qqgongyi.com',         // QQ Gongyi - Tencent's charitable foundation platform
	'www.ximalaya.com',         // Ximalaya - A Chinese online audio platform
	'www.dianping.com',         // Dianping - A Chinese online platform for finding and reviewing local businesses
	'www.suning.com',           // Suning - A leading Chinese online retailer
	'www.zhaopin.com',          // Zhaopin - A Chinese job recruitment platform
	'www.jianshu.com',          // Jianshu - A Chinese online writing platform
	'www.mafengwo.cn',          // Mafengwo - A Chinese travel information sharing platform
	'www.51cto.com',            // 51CTO - A Chinese IT technical community website
	'www.qidian.com',           // Qidian - A Chinese web novel platform
	'www.ctrip.com',            // Ctrip - A Chinese travel services provider
	'www.pconline.com.cn',      // PConline - A Chinese technology news and review website
	'www.cnzz.com',             // CNZZ - A Chinese web analytics service provider
	'www.telegraph.co.uk',      // The Telegraph - A British newspaper website	
	'www.ynet.com',             // Ynet - A Chinese news portal
	'www.ted.com',              // TED - A platform for ideas worth spreading
	'www.renren.com',           // Renren - A Chinese social networking service
	'www.pptv.com',             // PPTV - A Chinese online video streaming platform
	'www.liepin.com',           // Liepin - A Chinese online recruitment website
	'www.881903.com',           // 881903 - A Hong Kong radio station website
	'www.aipai.com',            // Aipai - A Chinese online video sharing platform
	'www.ttpaihang.com',        // Ttpaihang - A Chinese celebrity popularity ranking website
	'www.quyaoya.com',          // Quyaoya - A Chinese online ticketing platform
	'www.91.com',               // 91.com - A Chinese software download website
	'www.dianyou.cn',           // Dianyou - A Chinese game information website
	'www.tmtpost.com',          // TMTPost - A Chinese technology media platform
	'www.douban.com',           // Douban - A Chinese social networking service
	'www.guancha.cn',           // Guancha - A Chinese news and commentary website
	'www.so.com',               // So.com - A Chinese search engine
	'www.58.com',               // 58.com - A Chinese classified advertising website
	'www.google.com',           // Google - A multinational technology company
	'www.cnblogs.com',          // Cnblogs - A Chinese technology blog community
	'www.cntv.cn',              // CCTV - China Central Television official website
	'www.secoo.com',            // Secoo - A Chinese luxury e-commerce platform
];