Running an event loop within its own thread - python

I'm playing with Python's new(ish) asyncio stuff, trying to combine its event loop with traditional threading. I have written a class that runs the event loop in its own thread, to isolate it, and then provide a (synchronous) method that runs a coroutine on that loop and returns the result. (I realise this makes it a somewhat pointless example, because it necessarily serialises everything, but it's just as a proof-of-concept).
import asyncio
import aiohttp
from threading import Thread
class Fetcher(object):
def __init__(self):
self._loop = asyncio.new_event_loop()
# FIXME Do I need this? It works either way...
#asyncio.set_event_loop(self._loop)
self._session = aiohttp.ClientSession(loop=self._loop)
self._thread = Thread(target=self._loop.run_forever)
self._thread.start()
def __enter__(self):
return self
def __exit__(self, *e):
self._session.close()
self._loop.call_soon_threadsafe(self._loop.stop)
self._thread.join()
self._loop.close()
def __call__(self, url:str) -> str:
# FIXME Can I not get a future from some method of the loop?
future = asyncio.run_coroutine_threadsafe(self._get_response(url), self._loop)
return future.result()
async def _get_response(self, url:str) -> str:
async with self._session.get(url) as response:
assert response.status == 200
return await response.text()
if __name__ == "__main__":
with Fetcher() as fetcher:
while True:
x = input("> ")
if x.lower() == "exit":
break
try:
print(fetcher(x))
except Exception as e:
print(f"WTF? {e.__class__.__name__}")
To avoid this sounding too much like a "Code Review" question, what is the purpose of asynchio.set_event_loop and do I need it in the above? It works fine with and without. Moreover, is there a loop-level method to invoke a coroutine and return a future? It seems a bit odd to do this with a module level function.

You would need to use set_event_loop if you called get_event_loop anywhere and wanted it to return the loop created when you called new_event_loop.
From the docs
If there’s need to set this loop as the event loop for the current context, set_event_loop() must be called explicitly.
Since you do not call get_event_loop anywhere in your example, you can omit the call to set_event_loop.

I might be misinterpreting, but i think the comment by #dirn in the marked answer is incorrect in stating that get_event_loop works from a thread. See the following example:
import asyncio
import threading
async def hello():
print('started hello')
await asyncio.sleep(5)
print('finished hello')
def threaded_func():
el = asyncio.get_event_loop()
el.run_until_complete(hello())
thread = threading.Thread(target=threaded_func)
thread.start()
This produces the following error:
RuntimeError: There is no current event loop in thread 'Thread-1'.
It can be fixed by:
- el = asyncio.get_event_loop()
+ el = asyncio.new_event_loop()
The documentation also specifies that this trick (creating an eventloop by calling get_event_loop) only works on the main thread:
If there is no current event loop set in the current OS thread, the OS thread is main, and set_event_loop() has not yet been called, asyncio will create a new event loop and set it as the current one.
Finally, the docs also recommend to use get_running_loop instead of get_event_loop if you're on version 3.7 or higher

Related

Asyncio.sleep causes script to End Immediately

In my simple asyncio Python program below, bar_loop is supposed to run continuously with a 1 second delay between loops.
Things run as expected when we have simply
async def bar_loop(self):
while True:
print('bar')
However, when we add a asyncio.sleep(1), the loop will end instead of looping.
async def bar_loop(self):
while True:
print('bar')
await asyncio.sleep(1)
Why does asyncio.sleep() cause bar_loop to exit immediately? How can we let it loop with a 1 sec delay?
Full Example:
import asyncio
from typing import Optional
class Foo:
def __init__(self):
self.bar_loop_task: Optional[asyncio.Task] = None
async def start(self):
self.bar_loop_task = asyncio.create_task(self.bar_loop())
async def stop(self):
if self.bar_loop_task is not None:
self.bar_loop_task.cancel()
async def bar_loop(self):
while True:
print('bar')
await asyncio.sleep(1)
if __name__ == '__main__':
try:
foo = Foo()
asyncio.run(foo.start())
except KeyboardInterrupt:
asyncio.run(foo.stop())
Using Python 3.9.5 on Ubuntu 20.04.
This behavior has nothing to do with calling asyncio.sleep, but with the expected behavior of creating a task and doing nothing else.
Tasks will run in parallel in the the asyncio loop, while other code that uses just coroutine and await expressions can be thought as if run in a linear pattern - however, as the are "out of the way" of the - let's call it "visible path of execution", they also won't prevent that flow.
In this case, your program simply reaches the end of the start method, with nothing left being "awaited", the asyncio loop simply finishes its execution.
If you have no explicit code to run in parallel to bar_loop, just await for the task. Change your start method to read:
async def start(self):
self.bar_loop_task = asyncio.create_task(self.bar_loop())
try:
await self.bar_loop_task
except XXX:
# handle excptions that might have taken place inside the task

run expensive computation in another thread in blender

I'm developing a blender operator that needs to run an expensive for loop to generate a procedural animation. It looks roughly like this:
class MyOperator(bpy.types.Operator):
def execute(self, context):
data = []
for _ in range(10000):
data.append(run_something_expensive())
instantiate_data_into_blender(data)
return {"FINISHED"}
The problem of course is that when I run it from blender it takes a lot of time to finish and the UI becomes unresponsive. What is the recommended way to handle this? Is there a way to run this computation in another thread and potentially update the blender scene as the results get generated? (i.e. running instantiate_data_into_blender every once in a while as data becomes available)
Take a look at the asyncio module.
First: you have to import asyncio and define a loop.
import asyncio
loop = asyncio.get_event_loop()
Second: you create an async method in your operator.
async def massive_work(self, param):
returnedData = await your.longrun.func(param)
if returnedData == '':
return False
return True
Third: call the async method from your execute method
allDone = loop.run_until_complete(massive_work(dataToWorkOn))
if allDone != True:
self.report({'ERROR'}, "something went wrong!")
return {'CANCELLED'}
else:
self.report({'INFO'}, "all done!")
return {'FINISHED'}

await does not return the value of a future with the value set in a thread

This code should print "hi" 3 times, but it doesn't always print.
I made a gif that shows the code being executed:
from asyncio import get_event_loop, wait_for, new_event_loop
from threading import Thread
class A:
def __init__(self):
self.fut = None
def start(self):
"""
Expects a future to be created and puts "hi" as a result
"""
async def foo():
while True:
if self.fut:
self.fut.set_result('hi')
self.fut = None
new_event_loop().run_until_complete(foo())
async def make(self):
"""
Create a future and print your result when it runs out
"""
future = get_event_loop().create_future()
self.fut = future
print(await future)
a = A()
Thread(target=a.start).start()
for _ in range(3):
get_event_loop().run_until_complete(a.make())
This is caused by await future, because when I change
print(await future)
by
while not future.done():
pass
print(future.result())
the code always prints "hi" 3 times.
Is there anything in my code that causes this problem in await future?
Asyncio functions are not thread-safe, except where explicitly noted. For set_result to work from another thread, you'd need to call it through call_soon_threadsafe.
But in your case this wouldn't work because A.start creates a different event loop than the one the main thread executes. This creates issues because the futures created in one loop cannot be awaited in another one. Because of this, and also because there is just no need to create multiple event loops, you should pass the event loop instance to A.start and use it for your async needs.
But - when using the event loop from the main thread, A.start cannot call run_until_complete() because that would try to run an already running event loop. Instead, it must call asyncio.run_coroutine_threadsafe to submit the coroutine to the event loop running in the main thread. This will return a concurrent.futures.Future (not to be confused with an asyncio Future) whose result() method can be used to wait for it to execute and propagate the result or exception, just like run_until_complete() would have done. Since foo will now run in the same thread as the event loop, it can just call set_result without call_soon_threadsafe.
One final problem is that foo contains an infinite loop that doesn't await anything, which blocks the event loop. (Remember that asyncio is based on cooperative multitasking, and a coroutine that spins without awaiting doesn't cooperate.) To fix that, you can have foo monitor an event that gets triggered when a new future is available.
Applying the above to your code can look like this, which prints "hi" three times as desired:
import asyncio
from asyncio import get_event_loop
from threading import Thread
class A:
def __init__(self):
self.fut = None
self.have_fut = asyncio.Event()
def start(self, loop):
async def foo():
while True:
await self.have_fut.wait()
self.have_fut.clear()
if self.fut:
self.fut.set_result('hi')
self.fut = None
asyncio.run_coroutine_threadsafe(foo(), loop).result()
async def make(self):
future = get_event_loop().create_future()
self.fut = future
self.have_fut.set()
print(await future)
a = A()
Thread(target=a.start, args=(get_event_loop(),), daemon=True).start()
for _ in range(3):
get_event_loop().run_until_complete(a.make())

Python threading.Thread.join() is blocking

I'm working with asynchronous programming and wrote a small wrapper class for thread-safe execution of co-routines based on some ideas from this thread here: python asyncio, how to create and cancel tasks from another thread. After some debugging, I found that it hangs when calling the Thread class's join() function (I overrode it only for testing). Thinking I made a mistake, I basically copied the code that the OP said he used and tested it to find the same issue.
His mildly altered code:
import threading
import asyncio
from concurrent.futures import Future
import functools
class EventLoopOwner(threading.Thread):
class __Properties:
def __init__(self, loop, thread, evt_start):
self.loop = loop
self.thread = thread
self.evt_start = evt_start
def __init__(self):
threading.Thread.__init__(self)
self.__elo = self.__Properties(None, None, threading.Event())
def run(self):
self.__elo.loop = asyncio.new_event_loop()
asyncio.set_event_loop(self.__elo.loop)
self.__elo.thread = threading.current_thread()
self.__elo.loop.call_soon_threadsafe(self.__elo.evt_start.set)
self.__elo.loop.run_forever()
def stop(self):
self.__elo.loop.call_soon_threadsafe(self.__elo.loop.stop)
def _add_task(self, future, coro):
task = self.__elo.loop.create_task(coro)
future.set_result(task)
def add_task(self, coro):
self.__elo.evt_start.wait()
future = Future()
p = functools.partial(self._add_task, future, coro)
self.__elo.loop.call_soon_threadsafe(p)
return future.result() # block until result is available
def cancel(self, task):
self.__elo.loop.call_soon_threadsafe(task.cancel)
async def foo(i):
return 2 * i
async def main():
elo = EventLoopOwner()
elo.start()
task = elo.add_task(foo(10))
x = await task
print(x)
elo.stop(); print("Stopped")
elo.join(); print("Joined") # note: giving it a timeout does not fix it
if __name__ == "__main__":
loop = asyncio.new_event_loop()
asyncio.set_event_loop(loop)
assert isinstance(loop, asyncio.AbstractEventLoop)
try:
loop.run_until_complete(main())
finally:
loop.close()
About 50% of the time when I run it, It simply stalls and says "Stopped" but not "Joined". I've done some debugging and found that it is correlated to when the Task itself sent an exception. This doesn't happen every time, but since it occurs when I'm calling threading.Thread.join(), I have to assume it is related to the destruction of the loop. What could possibly be causing this?
The exception is simply: "cannot join current thread" which tells me that the .join() is sometimes being run on the thread from which I called it and sometimes from the ELO thread.
What is happening and how can I fix it?
I'm using Python 3.5.1 for this.
Note: This is not replicated on IDE One: http://ideone.com/0LO2D9

Is there a way to use asyncio.Queue in multiple threads?

Let's assume I have the following code:
import asyncio
import threading
queue = asyncio.Queue()
def threaded():
import time
while True:
time.sleep(2)
queue.put_nowait(time.time())
print(queue.qsize())
#asyncio.coroutine
def async():
while True:
time = yield from queue.get()
print(time)
loop = asyncio.get_event_loop()
asyncio.Task(async())
threading.Thread(target=threaded).start()
loop.run_forever()
The problem with this code is that the loop inside async coroutine is never finishing the first iteration, while queue size is increasing.
Why is this happening this way and what can I do to fix it?
I can't get rid of separate thread, because in my real code I use a separate thread to communicate with a serial device, and I haven't find a way to do that using asyncio.
asyncio.Queue is not thread-safe, so you can't use it directly from more than one thread. Instead, you can use janus, which is a third-party library that provides a thread-aware asyncio queue.
import asyncio
import threading
import janus
def threaded(squeue):
import time
while True:
time.sleep(2)
squeue.put_nowait(time.time())
print(squeue.qsize())
#asyncio.coroutine
def async_func(aqueue):
while True:
time = yield from aqueue.get()
print(time)
loop = asyncio.get_event_loop()
queue = janus.Queue(loop=loop)
asyncio.create_task(async_func(queue.async_q))
threading.Thread(target=threaded, args=(queue.sync_q,)).start()
loop.run_forever()
There is also aioprocessing (full-disclosure: I wrote it), which provides process-safe (and as a side-effect, thread-safe) queues as well, but that's overkill if you're not trying to use multiprocessing.
Edit
As pointed it out in other answers, for simple use-cases you can use loop.call_soon_threadsafe to add to the queue, as well.
If you do not want to use another library you can schedule a coroutine from the thread. Replacing the queue.put_nowait with the following works fine.
asyncio.run_coroutine_threadsafe(queue.put(time.time()), loop)
The variable loop represents the event loop in the main thread.
EDIT:
The reason why your async coroutine is not doing anything is that
the event loop never gives it a chance to do so. The queue object is
not threadsafe and if you dig through the cpython code you find that
this means that put_nowait wakes up consumers of the queue through
the use of a future with the call_soon method of the event loop. If
we could make it use call_soon_threadsafe it should work. The major
difference between call_soon and call_soon_threadsafe, however, is
that call_soon_threadsafe wakes up the event loop by calling loop._write_to_self() . So let's call it ourselves:
import asyncio
import threading
queue = asyncio.Queue()
def threaded():
import time
while True:
time.sleep(2)
queue.put_nowait(time.time())
queue._loop._write_to_self()
print(queue.qsize())
#asyncio.coroutine
def async():
while True:
time = yield from queue.get()
print(time)
loop = asyncio.get_event_loop()
asyncio.Task(async())
threading.Thread(target=threaded).start()
loop.run_forever()
Then, everything works as expected.
As for the threadsafe aspect of
accessing shared objects,asyncio.queue uses under the hood
collections.deque which has threadsafe append and popleft.
Maybe checking for queue not empty and popleft is not atomic, but if
you consume the queue only in one thread (the one of the event loop)
it could be fine.
The other proposed solutions, loop.call_soon_threadsafe from Huazuo
Gao's answer and my asyncio.run_coroutine_threadsafe are just doing
this, waking up the event loop.
BaseEventLoop.call_soon_threadsafe is at hand. See asyncio doc for detail.
Simply change your threaded() like this:
def threaded():
import time
while True:
time.sleep(1)
loop.call_soon_threadsafe(queue.put_nowait, time.time())
loop.call_soon_threadsafe(lambda: print(queue.qsize()))
Here's a sample output:
0
1443857763.3355968
0
1443857764.3368602
0
1443857765.338082
0
1443857766.3392274
0
1443857767.3403943
What about just using threading.Lock with asyncio.Queue?
class ThreadSafeAsyncFuture(asyncio.Future):
""" asyncio.Future is not thread-safe
https://stackoverflow.com/questions/33000200/asyncio-wait-for-event-from-other-thread
"""
def set_result(self, result):
func = super().set_result
call = lambda: func(result)
self._loop.call_soon_threadsafe(call) # Warning: self._loop is undocumented
class ThreadSafeAsyncQueue(queue.Queue):
""" asyncio.Queue is not thread-safe, threading.Queue is not awaitable
works only with one putter to unlimited-size queue and with several getters
TODO: add maxsize limits
TODO: make put corouitine
"""
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.lock = threading.Lock()
self.loop = asyncio.get_event_loop()
self.waiters = []
def put(self, item):
with self.lock:
if self.waiters:
self.waiters.pop(0).set_result(item)
else:
super().put(item)
async def get(self):
with self.lock:
if not self.empty():
return super().get()
else:
fut = ThreadSafeAsyncFuture()
self.waiters.append(fut)
result = await fut
return result
See also - asyncio: Wait for event from other thread

Categories