You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This raises an interesting question. Should this function create an event loop (possibly not if something else has already done this)? Should this function be taking an event_loop as an argument (like other functions do)? Guessing no since it is called in different processes. Should it take a function to create an event loop ( for example if we want to use uvloop#821 )? Or should it do something else (like try to use an existing event loop and only create one if it is absent)?
The text was updated successfully, but these errors were encountered:
I think the default should be try to use an existing event loop and only create one if it is absent but it would be great to also have an optional argument: function to create an event loop.
Currently
_worker_process
callsasyncio.get_event_loop()
ucx-py/ucp/utils.py
Line 128 in 9221fba
This function was deprecated in Python 3.10 ( https://bugs.python.org/issue39529 ). The docs recommend calling
asyncio.get_running_loop()
instead. Though this will raise aRuntimeError
if an event loop doesn't exist.This raises an interesting question. Should this function create an event loop (possibly not if something else has already done this)? Should this function be taking an
event_loop
as an argument (like other functions do)? Guessing no since it is called in different processes. Should it take a function to create an event loop ( for example if we want to useuvloop
#821 )? Or should it do something else (like try to use an existing event loop and only create one if it is absent)?The text was updated successfully, but these errors were encountered: