Asynchronous Programming#
Some I/O-bound and high-latency tasks can block the main thread and make your application unresponsive. To work around this, it's better to run these tasks independently of the main program flow. Asynchronous programming allows you to wait until a task is finished in the background while keeping your application responsive.
Ignis (and PyGObject in general) integrates with Python's asyncio module and provides convenient async/await syntax support.
Asynchronous functions in the Ignis documentation are prefixed with the async word before their name.
Calling Asynchronous Functions#
There are two common cases.
From a synchronous function
Use asyncio.create_task() to schedule execution.
import asyncio
from ignis.utils import Utils
asyncio.create_task(Utils.exec_sh_async("notify-send 'asynchrony!'"))
From another asynchronous function
Use the await keyword to wait for execution.
import asyncio
from ignis.utils import Utils
async def some_func() -> None:
await Utils.exec_sh_async("notify-send 'asynchrony!'")
# you still need create_task() because some_func is async
asyncio.create_task(some_func())
Choosing the right approach#
The best approach depends on your needs:
If you need to retrieve the function's output and/or control execution order, define a custom async function and use await.
Otherwise, if execution timing is not critical, use
asyncio.create_task()to run the task in the background.
Cancelling tasks#
Since asyncio.create_task() schedules execution for the future, you can cancel a task if needed.
task = asyncio.create_task(some_func())
# cancel task
task.cancel()