libev VS trio

Thank you for this.

I am interested in how concurrency can be represented elegantly and efficiently, so I am interested in how libraries can simplify async and make it easier to reason about and write

The libev and ioring support is great for IO scalability (https://github.com/enki/libev not sure if this is the official repo)

In Python I use the "select" module and use epoll on Linux.

I am currently thinking of designing an API that allows the registration of epoll-like listeners to arbitrary objects, including business objects, so you can efficiently register a listener on multiple behaviours of multiple arbitrary objects.

I wrote an async/await simulation in Java and my scheduler is really simple, it's just a for loop that checks to see if there are any tasks that can progress. I notice the switch_fiber in polyphony must do something similar. This is similar to a yield in a coroutine.

My async/await simulation takes the following program:

  task1:

Also, libev

It's done even now. See ev.h where they do this. Just that you disable GCC's aliasing warning. There's a reason why restrict and aliasing became important to deal with. It wasn't standards that killed it.

Incidentally, the alternative event loop implementation trio in python does not have "gather", you also need channels, and it's a deliberate design choice - there is some discussion about that in this ticket https://github.com/python-trio/trio/issues/2188

trio – a friendly Python library for async concurrency and I/O

trio. the best code, the best documentation, awesome community.

I'll +1 the Trio shoutout [1], but it's worth emphasizing that the core concept of Trio (nurseries) now exists in the stdlib in the form of task groups [2]. The article mentions this very briefly, but it's easy to miss, and I wouldn't describe it as a solution to this bug, anyways. Rather, it's more of a different way of writing multitasking code, which happens to make this class of bug impossible.

[1] https://github.com/python-trio/trio

[2] https://docs.python.org/3/library/asyncio-task.html#task-gro...

It's similar to manual memory management.

Structured concurrency is one approach to solving this problem. In a structured concurrency a promise would not go out of scope unhandled. Not sure how you would add APIs for it though.

See Python's trio nurseries idea which uses a python context manager.

https://github.com/python-trio/trio

I'm working on a syntax for state machines and it could be used as a DSL for promises. It looks similar to a bash pipeline but it matches predicates similar to prolog.

In theory you could wire up a tree of structured concurrency with this DSL.

https://github.com/samsquire/ideas4#558-assign-location-mult...

Not complete - doesn't include Task Groups [1]

In fairness they were only included in asyncio as of Python 3.11, which was released a couple of weeks ago.

These were an idea originally from Trio [2] where they're called "nurseries" instead of "task groups". My view is that you're better off using Trio, or at least anyio [3] which gives a Trio-like interface to asyncio. One particularly nice thing about Trio (and anyio) is that there's no way to spawn background tasks except to use task groups i.e. there's no analogue of asyncio's create_task() function. That is good because it guarantees that no task is ever left accidentally running in the background and no exception left silently uncaught.

[1] https://docs.python.org/3/library/asyncio-task.html#task-gro...

[2] https://github.com/python-trio/trio

[3] https://anyio.readthedocs.io/en/latest/