Skip to main content

A threading library.

Project description

https://github.com/eight04/pyWorker/actions/workflows/build.yml/badge.svg https://codecov.io/gh/eight04/pyWorker/branch/master/graph/badge.svg Documentation Status

A threading library written in python. Help you build threaded app.

This module was originally included in ComicCrawler.

Features

  • Pause, resume, stop, and restart a thread.

  • Support child threads.

  • Easily run asynchronous task across multiple threads.

  • Communicate between threads with events.

  • Use channel to broadcast events.

Install

pip install pythreadworker

Usage example

Basic operations and event:

#! python3

# Always use worker.sleep. pyWorker would process event queue during
# waiting.
from worker import create_worker, listen, sleep

@create_worker
def increaser():
  count = 1

  @listen("SET_VALUE")
  def _(event):
    nonlocal count
    # you don't need a lock to manipulate `count`
    count = event.data

  while True:
    print(count)
    # because the listener and the while loop are in the same thread
    count += 1
    sleep(1)

while True:
  command = input("input command: ")

  if command == "start":
    increaser.start()

  elif command == "stop":
    increaser.stop()

  elif command == "pause":
    increaser.pause()

  elif command == "resume":
    increaser.resume()

  elif command.startswith("set"):
    increaser.fire("SET_VALUE", int(command[4:]))

  elif command == "exit":
    increaser.stop()
    break

Async task:

#! python3

from worker import aynsc_, sleep

def long_work(t):
  sleep(t)
  return "Finished in {} second(s)".format(t)

# The async task will be executed in another thread.
pending = async_(long_work, 5)

# Do other stuff here...

# Wait the thread to complete and get the result. If the task is already
# finished, it returns directly with the result.
print(pending.get())

Use Channel to broadcast events:

#! python3

from worker import Worker, Channel

channel = Channel()

def create_printer(name):
  printer = Worker()

  @printer.listen("PRINT")
  def _(event):
    print(name, "recieved", event.data)

  channel.sub(printer)
  printer.start()
  return printer

foo = create_printer("foo")
bar = create_printer("bar")

channel.pub("PRINT", "Hello channel!")

foo.stop()
bar.stop()

Child thread and event bubbling/broadcasting:

#! python3

from worker import Worker, sleep

def create_thread(name, parent):
  thread = Worker(parent=parent)

  @thread.listen("HELLO")
  def _(event):
    print(name)

  return thread.start()

parent = create_thread("parent", None)
child = create_thread("child", parent)
grand = create_thread("grand", child)

# broadcast/bubble is happened in main thread. It doesn't gaurantee the
# execution order of listeners.
parent.fire("HELLO", broadcast=True)
sleep(1)
grand.fire("HELLO", bubble=True)
sleep(1)

# stop a parent thread would also stop its children
parent.stop()

How it works

The module creates a event queue for each thread, including the main thread. When blocking functions are called (worker.sleep, worker.wait_event, worker.Async.get, etc), they enter the event loop so the thread can process events, communicate with other threads, or raise an exception during the call.

Which also means that if you don’t use functions provided by pyThreadWorker, the module has no chance to affect your existing code. It should be easy to work with other frameworks.

API reference

http://pythreadworker.readthedocs.io/en/latest/

Notes

Changelog

  • 0.10.0 (Apr 19, 2022)

    • Change: require python 3.10+.

    • Change: now calling wait_* functions would initiate a root worker if there is no worker on the current thread.

  • 0.9.0 (Jun 8, 2018)

    • Change: The signature of `later()` is changed. You should use it to schedule a task on the specific thread.

    • Change: The listener registered by `listener()` shortcut would be removed once the thread is stopped.

    • Add: permanent and once arguments to Listener.

    • Add: Defer. A util to handle cross thread communication.

  • 0.8.0 (Mar 26, 2017)

    • Add print_traceback option to Worker.

    • Ability to use later as decorator.

    • Drop __all__ in __init__.py.

    • function rename: async -> async_, sync -> await_.

    • Async now extends Worker and needs start() to run.

    • later() now doesn’t use current thread as target by default. To use current thread as target, pass target=True.

    • Various function are able to used as decorator, including await_, async_, later.

    • Drop daemon Thread, use daemon Worker.

    • Add Worker.wait_until.

    • Add create_worker.

    • Refactor.

  • 0.7.0 (Feb 26, 2017)

    • Improve docs.

    • Drop def target(thread) syntax, use current() to get current thread instead.

    • Use pylint and sphinx.

    • Export more shortcuts.

  • 0.6.0 (Jul 1, 2016)

    • Add thread.later.

  • 0.5.1 (Apr 22, 2016)

    • Use float in sleep function.

  • 0.5.0 (Apr 22, 2016)

    • Add sync.

  • 0.4.0 (Apr 20, 2016) breaking change

    • Interface completely changed

    • Drop Message.put, .get

    • Drop UserWorker

    • Drop Worker.create_child. Use parent option in constructor instead.

    • Drop global_cleanup

    • Add sleep

    • Add current

    • Add Channel

    • Add Listener.priority

    • Add daemon option to Worker

    • Worker.cleanup –> Worker.update

    • Worker.message –> Worker.fire

    • Worker.wait_message –> Worker.wait_event

    • Worker.message_loop –> Worker.wait_forever

  • 0.3.0 (Jun 14, 2015)

    • Catch BaseException.

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

pythreadworker-0.10.0.tar.gz (12.2 kB view details)

Uploaded Source

Built Distribution

pythreadworker-0.10.0-py3-none-any.whl (11.9 kB view details)

Uploaded Python 3

File details

Details for the file pythreadworker-0.10.0.tar.gz.

File metadata

  • Download URL: pythreadworker-0.10.0.tar.gz
  • Upload date:
  • Size: 12.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.0 CPython/3.10.4

File hashes

Hashes for pythreadworker-0.10.0.tar.gz
Algorithm Hash digest
SHA256 438b983236888f9cb046aa131a791b6cf5843e4084e30b3fa863ae8821b382b0
MD5 28334d279763cc27cb24210788469bb0
BLAKE2b-256 9892ab9feafe1677a0b54156b1d1483c775e6bdad4f012d4bd28b52d96361ae3

See more details on using hashes here.

File details

Details for the file pythreadworker-0.10.0-py3-none-any.whl.

File metadata

File hashes

Hashes for pythreadworker-0.10.0-py3-none-any.whl
Algorithm Hash digest
SHA256 091b57b3d2ab10deb9eaf79ffa0482c13d36589582972061e8e9daeb285fb245
MD5 c5304e9ff6e6438a242a45e2901324f9
BLAKE2b-256 bd029c76be0b3397f32fd962d7a7be3c471735b4fa05661ef160e2aa2fd3811d

See more details on using hashes here.

Supported by

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