Skip to content

Workers

Workers run code that won't block the "main thread" controlling the user interface. If you block the main thread, your web page becomes annoyingly unresponsive. You should never block the main thread.

Happily, PyScript makes it very easy to use workers and uses a feature recently added to web standards called Atomics. You don't need to know about Atomics to use web workers, but it's useful to know that the underlying coincident library uses it under the hood.

Info

Sometimes you only need to await in the main thread on a method in a worker when neither window nor document are referenced in the code running on the worker.

In these cases, you don't need any special header or service worker as long as the method exposed from the worker returns a serializable result.

HTTP headers

To use the window and document objects from within a worker (i.e. use synchronous Atomics) you must ensure your web server enables the following headers (this is the default behavior for pyscript.com):

Access-Control-Allow-Origin: *
Cross-Origin-Opener-Policy: same-origin
Cross-Origin-Embedder-Policy: require-corp
Cross-Origin-Resource-Policy: cross-origin

If you're unable to configure your server's headers, you have two options:

  1. Use the mini-coi project to enforce headers.
  2. Use the service-worker attribute with the script element.

Option 1: mini-coi

For performance reasons, this is the preferred option so Atomics works at native speed.

The simplest way to use mini-coi is to copy the mini-coi.js file content and save it in the root of your website (i.e. /), and reference it as the first child tag in the <head> of your HTML documents:

<html>
  <head>
    <script src="/mini-coi.js"></script>
    <!-- etc -->
  </head>
  <!-- etc -->
</html>

Option 2: service-worker attribute

This allows you to slot in a custom service worker to handle requirements for synchronous operations.

Each <script type="m/py"> or <m/py-script> may optionally have a service-worker attribute pointing to a locally served file (the same way mini-coi.js needs to be served).

  • You can chose mini-coi.js itself or any other custom service worker, as long as it provides either the right headers to enable synchronous operations via Atomics, or it enables sabayon polyfill events.
  • Alternatively, you can copy and paste the sabayon Service Worker into your local project and point at that in the attribute. This will not change the original behavior of your project, it will not interfere with all default or pre-defined headers your application uses already but it will fallback to a (slower but working) synchronous operation that allows both window and document access in your worker logic.
<html>
  <head>
    <!-- PyScript link and script -->
  </head>
  <body>
    <script type="py" service-worker="./sw.js" worker>
      from pyscript import window, document

      document.body.append("Hello PyScript!")
    </script>
  </body>
</html>

Warning

Using sabayon as the fallback for synchronous operations via Atomics should be the last solution to consider. It is inevitably slower than using native Atomics.

If you must use sabayon, always reduce the amount of synchronous operations by caching references from the main thread.

# ❌ THIS IS UNNECESSARILY SLOWER
from pyscript import document

# add a data-test="not ideal attribute"
document.body.dataset.test = "not ideal"
# read a data-test attribute
print(document.body.dataset.test)

# - - - - - - - - - - - - - - - - - - - - -

# ✔️ THIS IS FINE
from pyscript import document

# if needed elsewhere, reach it once
body = document.body
dataset = body.dataset

# add a data-test="not ideal attribute"
dataset.test = "not ideal"
# read a data-test attribute
print(dataset.test)

In latter example the number of operations has been reduced from six to just four. The rule of thumb is: if you ever need a DOM reference more than once, cache it. 👍

Start working

To start your code in a worker, simply ensure the <script>, <py-script> or <mpy-script> tag pointing to the code you want to run has a worker attribute flag:

Evaluating code in a worker
<script type="py" src="./my-worker-code.py" worker></script>

You may also want to add a name attribute to the tag, so you can use pyscript.workers in the main thread to retrieve a reference to the worker:

<script type="py" src="./my-worker-code.py" worker name="my-worker"></script>
from pyscript import workers

my_worker = await workers["my-worker"]

Alternatively, to launch a worker from within Python running on the main thread use the pyscript.PyWorker class and you must reference both the target Python script and interpreter type:

Launch a worker from within Python
from pyscript import PyWorker

# The type MUST be given and can be either `micropython` or `pyodide`
my_worker = PyWorker("my-worker-code.py", type="micropython")

Worker interactions

Code running in the worker needs to be able to interact with code running in the main thread and perhaps have access to the web page. This is achieved via some helpful builtin APIs.

Note

For ease of use, the worker related functionality in PyScript is a simpler presentation of more sophisticated and powerful behaviour available via PolyScript.

If you are a confident advanced user, please consult the XWorker related documentation from the PolyScript project for how to make use of these features.

To synchronise serializable data between the worker and the main thread use the sync function in the worker to reference a function registered on the main thread:

Python code running on the main thread.
from pyscript import PyWorker

def hello(name="world"):
    return(f"Hello, {name}")

# Create the worker.
worker = PyWorker("./worker.py", type="micropython")

# Register the hello function as callable from the worker.
worker.sync.hello = hello
Python code in the resulting worker.
from pyscript import sync, window

greeting = sync.hello("PyScript")
window.console.log(greeting)

Alternatively, for the main thread to call functions in a worker, specify the functions in a __export__ list:

Python code on the worker.
import sys

def version():
    return sys.version

# Define what to export to the main thread.
__export__ = ["version", ]

Then ensure you have a reference to the worker in the main thread (for instance, by using the pyscript.workers):

Creating a named worker in the web page.
<script type="py" src="./my-worker-code.py" worker name="my-worker"></script>
Referencing and using the worker from the main thread.
from pyscript import workers

my_worker = await workers["my-worker"]

print(await my_worker.version())

The values passed between the main thread and the worker must be serializable. Try the example given above via this project on PyScript.com.

No matter if your code is running on the main thread or in a web worker, both the pyscript.window (representing the main thread's global window context) and pyscript.document (representing the web page's document object) will be available and work in the same way. As a result, a worker can reach into the DOM and access some window based APIs.

Warning

Access to the window and document objects is a powerful feature. Please remember that:

  • Arguments to and the results from such calls, when used in a worker, must be serializable, otherwise they won't work.
  • If you manipulate the DOM via the document object, and other workers or code on the main thread does so too, they may interfere with each other and produce unforeseen problematic results. Remember, with great power comes great responsibility... and we've given you a bazooka (so please remember not to shoot yourself in the foot with it).

Common Use Case

While it is possible to start a MicroPython or Pyodide worker from either MicroPython or Pyodide running on the main thread, the most common use case we have encountered is MicroPython on the main thread starting a Pyodide worker.

Here's how:

index.html

Evaluate main.py via MicroPython on the main thread
<!DOCTYPE html>
<html lang="en">
  <head>
    <meta charset="utf-8">
    <meta name="viewport" content="width=device-width,initial-scale=1">
    <!-- PyScript CSS -->
    <link rel="stylesheet" href="https://pyscript.net/releases/2024.10.2/core.css">
    <!-- This script tag bootstraps PyScript -->
    <script type="module" src="https://pyscript.net/releases/2024.10.2/core.js"></script>
    <title>PyWorker - mpy bootstrapping pyodide example</title>
    <script type="mpy" src="main.py"></script>
  </head>
</html>

main.py

MicroPython's main.py: bootstrapping a Pyodide worker.
from pyscript import PyWorker, document

# Bootstrap the Pyodide worker, with optional config too.
# The worker is:
#   * Owned by this script, no JS or Pyodide code in the same page can access
#     it.
#   * It allows pre-sync methods to be exposed.
#   * It has a ready Promise to await for when Pyodide is ready in the worker. 
#   * It allows the use of post-sync (methods exposed by Pyodide in the
#     worker).
worker = PyWorker("worker.py", type="pyodide")

# Expose a utility that can be immediately invoked in the worker. 
worker.sync.greetings = lambda: print("Pyodide bootstrapped")

print("before ready")
# Await until Pyodide has completed its bootstrap, and is ready.
await worker.ready
print("after ready")

# Await any exposed methods exposed via Pyodide in the worker.
result = await worker.sync.heavy_computation()
print(result)

# Show the result at the end of the body.
document.body.append(result)

# Free memory and get rid of everything in the worker.
worker.terminate()

worker.py

The worker.py script runs in the Pyodide worker.
from pyscript import sync

# Use any methods from main.py on the main thread.
sync.greetings()

# Expose any methods meant to be used from main.
sync.heavy_computation = lambda: 6 * 7

Save these files in a tmp folder, ensure your headers (just use npx mini-coi ./tmp to serve via localhost) then see the following outcome in the browser's devtools.

before ready
Pyodide bootstrapped
after ready
42