async_obj

A minimalist and lightweight Python wrapper to make any function -standalone or within an object- asynchronous.

If you are also tired of adding a whole bunch of code for every single case of simple threading, this small wrapper is my remedy. Requires >=Python 3.8.

async obj enables:

• Simply running a function in a dedicated thread.
• Check for completion of the function.
• Block until the function finishes.
• Receiving the returned result.
• In case of an exception, raising the exception on demand (whenever the result is requested).

Advantages:

• No compatibility concerns (e.g., asyncio), only depends on a few standard Python libraries.
• Essentially a 2 liner, no decorators or keywords.
• Mimics an object or a function and does not create a copy, hence minimum impact on memory and performance.

Installation

Easiest way is to use pip, only requirement is Python >=3.8.

pip install async_obj

Examples

• Run a function asynchronous and check for completion. Then collect the result.

from async_obj import async_obj
from time import sleep

def dummy_func(x:int):
    sleep(3)
    return x * x

#define the async version of the dummy function
async_dummy = async_obj(dummy_func)

print("Starting async function...")
async_dummy(2)  # Run dummy_func asynchronously
print("Started.")

while True:
    print("Checking whether the async function is done...")
    if async_dummy.async_obj_is_done():
        print("Async function is done!")
        print("Result: ", async_dummy.async_obj_get_result(), " Expected Result: 4")
        break
    else:
        print("Async function is still running...")
        sleep(1)

---

• Alternatively, block until the function is completed, also retrieve any results.

print("Starting async function...")
async_dummy(4)  # Run dummy_func asynchronously
print("Started.")
print("Blocking until the function finishes...")
result = async_dummy.async_obj_wait()
print("Function finished.")
print("Result: ", result, " Expected Result: 16")

---

• Raise propagated exceptions, whenever the result is requested either with async_obj_get_result() or with async_obj_wait().

print("Starting async function with an exception being expected...")
async_dummy(None) # pass an invalid argument to raise an exception
print("Started.")
print("Blocking until the function finishes...")
try:
    result = async_dummy.async_obj_wait()
except Exception as e:
    print("Function finished with an exception: ", str(e))
else:
    print("Function finished without an exception, which is unexpected.")

---

• Same functionalities are available for functions within class instances.

class dummy_class:
    x = None

    def __init__(self):
        self.x = 5

    def dummy_func(self, y:int):
        sleep(3)
        return self.x * y

dummy_instance = dummy_class()
#define the async version of the dummy function within the dummy class instance
async_dummy = async_obj(dummy_instance)

print("Starting async function...")
async_dummy.dummy_func(4)  # Run dummy_func asynchronously
print("Started.")
print("Blocking until the function finishes...")
result = async_dummy.async_obj_wait()
print("Function finished.")
print("Result: ", result, " Expected Result: 20")

To Do

1. Providing same concurrency functionalities for getter, setter and similar extension.
2. Improve autocompletion support for better IDE convenience.