Skip to main content

Actor model for Python

Project description

splut

Actor model for Python

Usage example

>>> from concurrent.futures import ThreadPoolExecutor
>>> from splut.actor import Join, Spawn
>>> from urllib.request import urlopen
>>> import sys
>>>
>>> class Pipe:
...
...     traffic = 0
...
...     def __init__(self, index):
...         self.index = index
...
...     def fetch(self, url):
...         def log(message):
...             print(f"[{self.index}] [{url}] {message}", file = sys.stderr) # Demonstrate worker (and thus thread) utilisation.
...         log('Fetch.')
...         with urlopen(url) as f:
...             data = f.read()
...         n = len(data)
...         log(f"Got: {n}")
...         self.traffic += n # Effectively self is locked, so this is thread-safe.
...         return data
...
>>> class Sum:
...
...     def __init__(self, pipeactor):
...         self.pipeactor = pipeactor
...
...     async def bytecount(self, dotcoms):
...         futures = [self.pipeactor.fetch(f"https://www.{dotcom}.com/") for dotcom in dotcoms] # Collect futures eagerly.
...         return sum(map(len, await Join(futures)))
...
>>> pipes = [Pipe(i) for i in range(3)]
>>> with ThreadPoolExecutor() as e:
...     spawn = Spawn(e)
...     pipeactor = spawn(*pipes) # One mailbox, multiple independent workers.
...     sumactor = spawn(Sum(pipeactor)) # An actor is cheap to create, unlike a thread.
...     bytecount = sumactor.bytecount(['facebook', 'google', 'tumblr', 'yahoo', 'youtube']).wait()

Total bytes piped by workers matches the total we have:
>>> sum(p.traffic for p in pipes) == bytecount
True

API

splut.actor

Spawn Objects

class Spawn()

__init__
def __init__(executor)

Spawned actors will use threads from the given executor.

__call__
def __call__(*objs)

Create an actor backed by the given worker object(s), each of which is used in a single-threaded way. Calling a method on the returned actor returns a Future immediately, which eventually becomes done with the result of a worker method of the same name (or never if the worker method hangs). A worker method may be async, in which case it can await futures returned by other actors, releasing the worker in the meantime.

Join Objects

class Join()

Make multiple futures awaitable as a unit. In the zero futures case this resolves (to an empty list) without suspending execution. Otherwise if any future hangs, so does this. Otherwise if any future failed, all such exceptions are raised as a chain. Otherwise all results are returned as a list.

splut.actor.future

Future Objects

class Future()

wait
def wait()

Block until there is an outcome, then return/raise it. For use outside actors, or within one if you know the future is done and don't want to suspend execution with await in that case.

andforget
def andforget(log)

Send any exception to the given log.

splut.bg

Sleeper Objects

class Sleeper()

interrupt
def interrupt()

If a sleep is in progress that sleep returns now, otherwise the next sleep will return immediately. This is similar behaviour to interrupting a maybe-sleeping thread in Java.

splut.bg.delay

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

splut-17.tar.gz (7.5 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

splut-17-py3-none-any.whl (7.7 kB view details)

Uploaded Python 3

File details

Details for the file splut-17.tar.gz.

File metadata

  • Download URL: splut-17.tar.gz
  • Upload date:
  • Size: 7.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.10.12

File hashes

Hashes for splut-17.tar.gz
Algorithm Hash digest
SHA256 42fe248d1ad8db054a28de99d17319c22a836295bf4847b8544e985bbdbc67af
MD5 1a6f3067c399980e7ee5382c4943873f
BLAKE2b-256 19fcc3ebeac4e2156defc74e1315e995ae82029d3ce241ad9801c28ecdc0b305

See more details on using hashes here.

File details

Details for the file splut-17-py3-none-any.whl.

File metadata

  • Download URL: splut-17-py3-none-any.whl
  • Upload date:
  • Size: 7.7 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.10.12

File hashes

Hashes for splut-17-py3-none-any.whl
Algorithm Hash digest
SHA256 fb6023c778dfca22c2b95d7eecfeaa5df948406617f875d303a69f9ea4d59f9d
MD5 b3d1a821e06ec601053afe7c21e78426
BLAKE2b-256 a86f6e12af325dc56e45227ddeb61b8762a69ab9a1bade51d968bee234fb836b

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page