ZetCode

Python __await__ Method

Last modified April 8, 2025

This comprehensive guide explores Python's __await__ method, the special method that makes objects awaitable. We'll cover async programming, coroutines, awaitable objects, and practical examples.

Basic Definitions

The __await__ method returns an iterator that is used to implement awaitable objects. It allows an object to be used with the await expression in async functions.

Key characteristics: it must return an iterator, is used by the async/await syntax, and enables custom awaitable objects. It's part of Python's async programming model introduced in Python 3.5.

Basic __await__ Implementation

Here's a simple implementation showing how __await__ makes a custom object awaitable. The method must return an iterator.

basic_await.py
class SimpleAwaitable:
    def __await__(self):
        yield
        return "Done"

async def main():
    result = await SimpleAwaitable()
    print(result)  # Output: Done

import asyncio
asyncio.run(main())

This example creates a basic awaitable object. The __await__ method yields once and returns a value. When awaited, it pauses execution until the yield completes.

The yield is crucial - it makes the method a generator, which is required for __await__. The returned value becomes the await expression's result.

Awaitable with Asynchronous Operation

This example shows a more practical awaitable that simulates an async operation with a delay.

async_operation.py
class AsyncOperation:
    def __init__(self, delay):
        self.delay = delay
    
    def __await__(self):
        yield from asyncio.sleep(self.delay)
        return f"Completed after {self.delay}s"

async def main():
    op = AsyncOperation(1.5)
    print("Starting operation")
    result = await op
    print(result)

asyncio.run(main())

This awaitable object uses asyncio.sleep to simulate an async operation. The yield from delegates to another awaitable.

When awaited, it pauses execution for the specified delay before returning the completion message. This pattern is common for wrapping async operations.

Custom Future Implementation

This example demonstrates implementing a basic Future-like object with __await__ that can be set with a result later.

custom_future.py
class CustomFuture:
    def __init__(self):
        self._result = None
        self._done = False
    
    def set_result(self, result):
        self._result = result
        self._done = True
    
    def __await__(self):
        while not self._done:
            yield
        return self._result

async def set_future(future, delay, value):
    await asyncio.sleep(delay)
    future.set_result(value)

async def main():
    future = CustomFuture()
    asyncio.create_task(set_future(future, 2, "Future result"))
    print("Waiting for future...")
    result = await future
    print(f"Got: {result}")

asyncio.run(main())

This CustomFuture can be awaited until its result is set. The __await__ method yields until _done becomes True, then returns the result.

This mimics how real Futures work in asyncio, allowing decoupled result setting and awaiting. The pattern is useful for bridging callback-based APIs with async.

Chaining Awaitables

This example shows how to chain multiple awaitable objects together using __await__.

chaining_awaitables.py
class AddAwaitable:
    def __init__(self, a, b):
        self.a = a
        self.b = b
    
    def __await__(self):
        result = yield from self.a.__await__()
        result += yield from self.b.__await__()
        return result

async def slow_add(a, b):
    await asyncio.sleep(1)
    return a + b

async def main():
    a = slow_add(10, 20)
    b = slow_add(30, 40)
    combined = AddAwaitable(a, b)
    result = await combined
    print(f"Total: {result}")  # Output: Total: 100

asyncio.run(main())

The AddAwaitable class chains two awaitables, summing their results. It uses yield from to await each operand sequentially.

This demonstrates how __await__ can compose multiple async operations into new awaitable objects. The pattern is useful for building complex async workflows.

Async Context Manager with __await__

This example implements an async context manager using __await__ for resource management.

async_context.py
class AsyncResource:
    def __init__(self, name):
        self.name = name
    
    async def __aenter__(self):
        print(f"Opening {self.name}")
        await asyncio.sleep(0.5)
        return self
    
    async def __aexit__(self, *args):
        print(f"Closing {self.name}")
        await asyncio.sleep(0.5)
    
    def __await__(self):
        return self.__aenter__().__await__()

async def main():
    async with AsyncResource("DB Connection") as resource:
        print(f"Using {resource.name}")
        await asyncio.sleep(1)
    
    # Alternative using await directly
    resource = await AsyncResource("File")
    try:
        print(f"Using {resource.name}")
        await asyncio.sleep(1)
    finally:
        await resource.__aexit__(None, None, None)

asyncio.run(main())

This shows two ways to use an async resource: with async with and directly with await. The __await__ method delegates to __aenter__.

The pattern is useful when you need both context manager and direct await support. It demonstrates how __await__ can integrate with other async protocols.

Best Practices

Source References

Author

My name is Jan Bodnar, and I am a passionate programmer with extensive programming experience. I have been writing programming articles since 2007. To date, I have authored over 1,400 articles and 8 e-books. I possess more than ten years of experience in teaching programming.

List all Python tutorials.