Skip to main content

Call and interop Node.js APIs with Python

Project description


NPM version PyPI Build Status Gitpod ready-to-code

Interoperate Node.js and Python. Work in progress.

Requires Node.js 16 and Python 3.8 or newer.

Key Features

  • Ability to call async and sync functions and get object properties with a native feel
  • Built-in garbage collection
  • Bidirectional callbacks with arbitrary arguments
  • Iteration and exception handling support
  • Object inspection allows you to easily console.log or print() any foreign objects
  • (Bridge to call Python from JS) Python class extension and inheritance. See pytorch and tensorflow examples.
  • (Bridge to call JS from Python) Native decorator-based event emitter support

Basic usage example

See some examples here. See documentation below and in here.

Access JavaScript from Python

pip3 install javascript
from javascript import require, globalThis

chalk, fs = require("chalk"), require("fs")

print("Hello","world!"), "it's", globalThis.Date().toLocaleString())
fs.writeFileSync("HelloWorld.txt", "hi!")

Access Python from JavaScript

Make sure to have the dependencies installed before hand!

npm i pythonia
import { python } from 'pythonia'
// Import tkinter
const tk = await python('tkinter')
// All Python API access must be prefixed with await
const root = await tk.Tk()
// A function call with a $ suffix will treat the last argument as a kwarg dict
const a = await tk.Label$(root, { text: 'Hello World' })
await a.pack()
await root.mainloop()
python.exit() // Make sure to exit Python in the end to allow node to exit. You can also use process.exit.


Gitpod ready-to-code

Check out some cool examples below! Try them on Gitpod! Click the Open in Gitpod link above, and then open the examples folder.

PyTorch numpy tensorflow mineflayer

Bridge feature comparison

Unlike other bridges, you may notice you're not just writing Python code in JavaScript, or vice-versa. You can operate on objects on the other side of the bridge as if the objects existed on your side. This is achieved through real interop support: you can call callbacks, and do loss-less function calls with any arguments you like (with the exception of floating points percision of course).

python(ia) bridge javascript bridge npm:python-bridge
Garbage collection
Class extension support Not built-in (rare use case), can be manually done with custom proxy
Passthrough stdin ❌ (Standard input is not piped to bridge processes. Instead, listen to standard input then expose an API on the other side of the bridge recieve the data.)
Passthrough stdout, stderr
Long-running sync calls
Long-running async calls ❌ (need to manually create new thread) ✔ (AsyncTask) ❌ (need to manually create new thread)
Call classes
Inline eval
Dependency Management
Local File Imports
Error Management
Object inspection

Notable details

  • The ffid keyword is reserved. You cannot use it in variable names, object keys or values as this is used to internlly track objects.
  • On the bridge to call JavaScript from Python, due to the limiatations of Python and cross-platform IPC, we currently communicate over STDERR which means that JSON output in JS STDERR can interfere with the bridge. The same issue exists on Windows with pythoni. You are very unlikely to have issues with this, but it will be fixed soon.


This bridge works through standard input/output pipes, there are no native modules and the communication can happen through anywhere--either pipes or network sockets.

You can import and call any JS or Python class you want, with few exceptions.

How it works. For every property access, there is a communication protocol where one side may access properties on the other, and also complete function calls. Non-primitive values are sent as foreign object reference IDs (FFID). These FFIDs exist in a map on both sides of the bridge, and map numeric IDs with a object reference.

On the opposite side to the one which holds a reference, this FFID is assigned to a Proxy object. In JS, a ES6 proxy is used, and in Python, the proxy is a normal class with custom __getattr__ and other magic methods. Each proxy property access is mirrored on the other side of the bridge.

Proxy objects on both sides of the bridge are GC tracked. In JavaScript, all python Proxy objects are registered to a FinalizationRegistry. In Python, __del__ is used to track the Proxy object's destruction. When the proxy object is destoryed on one side of the bridge, its refrence is removed from the other side of the bridge. This means you don't have to deal with memory management.

  • When doing a function call, any foreign objects will be sent to you as a reference. For example, if you're in JavaScript and do a function call to Python that returns an array, you won't get a JS array back, but you will get a reference to the Python array. You can still access the array normally with the [] notation, as long as you use await. If you would like the bridge to turn the foreign refrence to something native, you can request a primitive value by calling .valueOf() on the Python array. This would give you a JS array. It works the same the other way around.
  • The above behavior makes it very fast to pipe data from one function onto another, avoiding costly conversions.
  • This above behavior is not present for callbacks and function parameters. The bridge will try to serialize what it can, and will give you a foreign reference if it's unable to serialize something. So if you pass a JS object, you'll get a Python dict, but if the dict contains something like a class, you'll get a reference in its place.


You can import the bridge module with

from javascript import require

This will import the require function which you can use just like in Node.js. This is a slightly modified require function which does dependency management for you. The first paramater is the name or location of the file to import. Internally, this calls the ES6 dynamic import() function. Which supports both CommonJS and ES6 modules.

If you are passing a module name (does not start with / or include a .) such as 'chalk', it will search for the dependency in the internal node_module folder and if not found, install it automatically. This install will only happen once, it won't impact startup afterwards.

The second paramater to the built-in require function is the version of the package you want, for example require('chalk', '^3') to get a version greater than major version 3. Just like you would if you were using npm install. It's reccomended to only use the major version as the name and version will be internally treated as a unique package, for example 'chalk--^3'. If you leave this empty, we will install latest version instead, or use the version that may already be installed globally.


  • All function calls to JavaScript are thread synchronous
  • ES6 classes can be constructed without new
  • ES5 classes can be constructed with the .new psuedo method
  • Use @On decorator when binding event listeners. Use off() to disable it.
  • All callbacks run on a dedicated callback thread. DO NOT BLOCK in a callback or all other events will be blocked. Instead:
  • Use the @AsyncTask decorator when you need to spawn a new thread for an async JS task.

For more, see docs/

Basic import

Let's say we have a file in JS like this called time.js ...

function whatTimeIsIt() {
    return (new Date()).toLocaleString()
module.exports = { whatTimeIsIt }

Then we can call it from Python !

from javascript import require
time = require('./time.js')

Event emitter

You must use the provided On, Once, decorator and off function over the normal dot methods.


const { EventEmitter } = require('events')
class MyEmitter extends EventEmitter {
    counter = 0
    inc() {
        this.emit('increment', ++this.counter)
module.exports = { MyEmitter }

from javascript import require, On, off
MyEmitter = require('./emitter.js')
# New class instance
myEmitter = MyEmitter()
# Decorator usage
@On(myEmitter, 'increment')
def handleIncrement(this, counter):
    print("Incremented", counter)
    # Stop listening. `this` is the this variable in JS.
    off(myEmitter, 'increment', handleIncrement)
# Trigger the event handler

ES5 class


function MyClass(num) {
    this.getNum = () => num
module.exports = { MyClass }

MyEmitter = require('./es5.js')
myClass =



module.exports = { items: [5, 6, 7, 8] }

items = require('./items.js')
for item in items:



export function method(cb, salt) {
    cb(42 + salt)

method = require('./callback').method
# Example with a lambda, but you can also pass a function ref
method(lambda v: print(v), 2) # Prints 44


The magic behind this is the usage of Proxy chains which permits call stack build up, until a .then call for property access or a function call is done. Afterwards, the callstack is sent and executed in Python.

  • All the Python APIs are async. You must await them all.
  • Use python.exit() or process.exit() at the end to quit the Python process.
  • This library doesn't manage the packaging.
    • Right now you need to install all the deps from pip globally, but later on we may allow loading from pip-envs.
  • When you do a normal Python function call, you can supply "positional" arguments, which must be in the correct order to what the Python function expects.
  • Some Python objects accept arbitrary keyword arguments. You can call these functions by using the special $ function syntax.
    • When you do a function call with a $ before the parenthesis, such as await some.pythonCall$(), the final argument is evaluated as a kwarg dictionary. You can supply named arguments this way.
  • Property access with a $ at the end acts as a error suppression operator.
    • Any errors will be ignored and instead undefined will be returned
  • See docs/ for more docs, and the examples for more info


Basic import

Let's say we have a file in Python like this called ...

import datetime
def what_time_is_it():
  return str(

Then we can call it from JavaScript !

import { python } from 'pythonia'
const time = await python('./')
console.log("It's", await time.what_time_is_it())


  • When iterating a Python object, you must use a for await loop instead of a normal for-of loop.

import os
def get_files():
  for f in os.listdir():
    yield f


const iter = await python('./')
const files = await iter.get_files()
for await (const file of files) {

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for javascript, version 1!0.2.9
Filename, size File type Python version Upload date Hashes
Filename, size javascript-1!0.2.9-py3-none-any.whl (31.8 kB) File type Wheel Python version py3 Upload date Hashes View
Filename, size javascript-1!0.2.9.tar.gz (36.8 kB) File type Source Python version None Upload date Hashes View

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring DigiCert DigiCert EV certificate Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page