JavaScript Wisp Implementation

This is an implementation of a Wisp client and server, written in Javascript for use in NodeJS and in the browser.

Library Entrypoints:

• @mercuryworkshop/wisp-js/client - Only contains the client code.
• @mercuryworkshop/wisp-js/server - Contains the server code, and the logging module.
• @mercuryworkshop/wisp-js - Contains the server, client, and logging code.

All of these entrypoints support being imported as either a CommonJS or ES6 module.

Server CLI:

THere is a CLI interface available for the Wisp server, and it can be used by installing the package with npm, then running:

$ npx wisp-js-server --help
Usage: wisp-js-server [options]

A Wisp server implementation written in Javascript. (v0.3.1)

Options:
  -V, --version                 output the version number
  -H, --host <host>             The hostname the server will listen on. (default: "127.0.0.1")
  -P, --port <port>             The port number to run the server on. (default: 5001)
  -L, --logging <log_level>     The log level to use. This is either DEBUG, INFO, WARN, ERROR, or
                                NONE. (default: "INFO")
  -S, --static <static_dir>     The directory to serve static files from. (optional)
  -C, --config <config_path>    The path to your Wisp server config file. This is the same format as
                                `wisp.options` in the API. (optional)
  -O, --options <options_json>  A JSON string to set the Wisp config without using a file.
                                (optional)
  -h, --help                    display help for command

You may also clone this repository and run npn run server_cli -- --help.

The config file is a JSON file with the same entries as the global server config in the API.

Client API:

Importing the Client Library:

To use the library as an ES6 module, either in Node or using a bundler for the browser, include the following import:

import { client as wisp } from "@mercuryworkshop/wisp-js/client";

To use it in Node with CommonJS:

const { client: wisp } = require("@mercuryworkshop/wisp-js/client");

If you are not using a bundler, you may import the files in the dist folder of the package. The wisp-client.mjs file is an ES6 module that has the same entrypoint as the example above. The wisp-client.js file is a regular JS file that produce a global variable named wisp_client, which contains all of the exported modules.

Connecting to a Wisp Server:

You can create a new Wisp connection by creating a new ClientConnection object. The only argument to the constructor is the URL of the Wisp websocket server. Use the open event to know when the Wisp connection is ready.

import { client as wisp } from "@mercuryworkshop/wisp-js/client";

let conn = new wisp.ClientConnection("wss://example.com/wisp/");
conn.onopen = () => {
  console.log("wisp connection is ready!");
};
conn.onclose = () => {
  console.log("wisp connection closed");
};
conn.onerror = () => {
  console.log("wisp connection error");
};

Creating New Streams:

Once you have your WispConnection object, and you have waited for the connection to be established, you can use the WispConnection.create_stream method to create new streams. The two arguments to this function are the hostname and port of the new stream, and a WispStream object will be returned. You can also pass a third argument to create_stream, which is the type of the stream, and it can be either "tcp" (the default) or "udp".

For receiving incoming messages, use the onmessage callback on the WispStream object. The returned data will always be a Uint8Array. The onclose callback can be used to know when the stream is closed.

You can use stream.send to send data to the stream, and it must take a Uint8Array as the argument. Newly created streams are available for writing immediately, so you don't have to worry about waiting to send your data.

let stream = conn.create_stream("www.google.com", 80);
stream.onmessage = (data) => {
  let text = new TextDecoder().decode(data);
  console.log(text);
};
stream.onclose = (reason) => {
  console.log("stream closed for reason: " + reason);
};

let payload = "GET / HTTP/1.1\r\nHost: www.google.com\r\nConnection: close\r\n\r\n";
stream.send(new TextEncoder().encode(payload));

Using the WebSocket Polyfill:

The polyfill.js file provides an API similar to the regular DOM WebSocket API. Instead of creating new WebSocket objects, create WispWebSocket objects. Make sure the URL ends with the hostname and port you want to connect to. If you have code that uses the older wsproxy protocol, you can use this polyfill to provide Wisp support easily.

import { client as wisp } from "@mercuryworkshop/wisp-js/client";

let ws = new wisp.WispWebSocket("wss://example.com/ws/alicesworld.tech:80");
ws.binaryType = "arraybuffer";
ws.addEventListener("open", () => {
  let payload = "GET / HTTP/1.1\r\nHost: alicesworld.tech\r\nConnection: keepalive\r\n\r\n";
  ws.send(payload);
});
ws.addEventListener("message", (event) => {
  let text = new TextDecoder().decode(event.data);
  console.log("message from stream 1: ", text.slice(0, 128));
});

The wisp_client._wisp_connections object will be used to manage the active Wisp connections. This object is able to store multiple active Wisp connections, identified by the websocket URL.

Server API:

Importing the Server Library:

To use the library as an ES6 module, either in Node or using a bundler for the browser, include the following import:

import { server as wisp } from "@mercuryworkshop/wisp-js/server";

To use it in Node with CommonJS:

const { server: wisp } = require("@mercuryworkshop/wisp-js/client");

This is designed to be a drop-in replacement for wisp-server-node. You can replace your old import with one of the above examples, and your application will still work in the same way.

Basic Example:

This example uses the node:http module as a basic web server. It accepts new Wisp connections from incoming websockets.

import { server as wisp } from "@mercuryworkshop/wisp-js/server";
import http from "node:http";

const server = http.createServer((req, res) => {
  res.writeHead(200, { "Content-Type": "text/plain" });
  res.end("wisp server js rewrite");
});

server.on("upgrade", (req, socket, head) => {
  wisp.routeRequest(req, socket, head);
});

server.on("listening", () => {
  console.log("HTTP server listening");
});

server.listen(5001, "127.0.0.1");

Example With Express:

import { server as wisp } from "@mercuryworkshop/wisp-js/server";
import express from "express";
import morgan from "morgan";

const app = express();
const port = process.env.PORT || 5001;

app.use(morgan("combined"));
app.use(express.static("./"));

const server = app.listen(port, () => {
  console.log("Listening on port: ", port)
});

server.on("upgrade", (request, socket, head) => {
  wisp.routeRequest(request, socket, head);
});

Change the Log Level:

By default, all info messages are shown. You can change this by importing logging from the module, and calling logging.set_level to set it to one of the following values:

• logging.DEBUG
• logging.INFO (default)
• logging.WARN
• logging.ERROR
• logging.NONE

import { server as wisp, logging } from "@mercuryworkshop/wisp-js/server";

logging.set_level(logging.DEBUG);

Changing Server Settings:

To change settings globally for the Wisp server, you can use the wisp.options object. Here is a list of all of the available settings:

Blacklist / Whitelist Options:

• options.hostname_blacklist - An array of regex objects to match against the destination server. Any matches will be blocked.
• options.hostname_whitelist - Same as hostname_blacklist, but only matches will be allowed through, and setting this will supersede hostname_blacklist.
• options.port_blacklist - An array of port numbers or ranges to block on the destination server. Specific ports are expressed as a single number, and ranges consist of a two element array containing the start and end. For example 80 and [3000, 4000] are both valid entries in this array.
• options.port_whitelist - Same as port_whitelist, but only matches will be allowed through, and setting this will supersede port_blacklist.

Stream Restrictions:

• options.stream_limit_per_host - The maximum number of streams that may be open to a single hostname, per connection. Defaults to no limit.
• options.stream_limit_total - The total number of streams that may be open to all hosts combined, per connection. Defaults to no limit.
• options.allow_udp_streams - If this is false, UDP streams will be blocked. Defaults to true.
• options.allow_tcp_streams - If this is false, TCP streams will be blocked. Defaults to true.

IP Restrictions:

• options.allow_direct_ip - Allow connections directly to IP addresses, which bypasses the server-side DNS resolution. Turning this off allows the server administrator to enforce a block list more effectively. Defaults to true.
• options.allow_private_ips - Allow connections to private IP addresses. Defaults to false.
• options.allow_loopback_ips - Allow connections to the server's localhost (127.0.0.1) and other loopback IPs. Defaults to false.
• options.parse_real_ip - Parse the client's real IP from the X-Forwarded-For and X-Real-IP headers. Defaults to true.
• options.parse_real_ip_from - A list of IP addresses to allow parsing the real IP from. Defaults to ["127.0.0.1"].

DNS Settings:

• options.dns_ttl - The time to live for cached DNS responses, in seconds. Defaults to 120 seconds.
• options.dns_method - The method to use for DNS resolution. This is either "lookup", which uses the system DNS, or "resolve", which uses the Node dns.resolve functions. This may also be a custom async function, which takes the hostname as its only argument and returns the resolved IP address. Defaults to "lookup".
• options.dns_servers - A list of strings containing IP addresses for custom DNS servers. This is only used if dns_method is set to "resolve". By default, this is unset, and DNS queries will use the system DNS servers.
• options.dns_result_order - Controls whether or not IPv4 or IPv6 addresses are prioritized. This can be either "ipv4first", "ipv6first", or "verbatim". "verbatim" uses the original order that the system DNS returns the results in, and only has special meaning if the DNS method is "lookup". If the DNS method is "resolve", "verbatim" is treated the same as "ipv6first". Defaults to "verbatim".

For example:

wisp.options.port_whitelist = [
  [5000, 6000],
  80,
  443
]
wisp.options.hostname_blacklist = [
  /google\.com/,
  /reddit\.com/,
]

wisp.options.dns_method = "resolve";
wisp.options.dns_servers = ["1.1.1.3", "1.0.0.3"];
wisp.options.dns_result_order = "ipv4first";

Copyright:

This library is licensed under the GNU AGPL v3.

Copyright Notice:

wisp-js: a Wisp client implementation written in JavaScript
Copyright (C) 2024 Mercury Workshop

This program is free software: you can redistribute it and/or modify
it under the terms of the GNU Affero General Public License as published
by the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.

This program 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 Affero General Public License for more details.

You should have received a copy of the GNU Affero General Public License
along with this program.  If not, see <https://www.gnu.org/licenses/>.