Skip to content

ichuan/simple_threadpool

Repository files navigation

simple_threadpool

A simple threadpool, supports large iterable job queue(add on demand, compared to concurrent.futures.ThreadPoolExecutor and multiprocessing.pool.ThreadPool).

Install

pip install simple_threadpool

Usage

from __future__ import print_function

import random, time
from threading import current_thread
from simple_threadpool import ThreadPool


def my_worker(arg):
    '''
    custom worker
    '''
    print('%s: ' % current_thread().name, arg + 1)
    time.sleep(random.random())


def large_iterable(size):
    for i in range(size):
        print('getting %s' % i)
        yield i


def callback(result):
    print(result)


# create a ThreadPool instance with 2 threads
tp = ThreadPool(my_worker, max_workers=2)
print('max_workers: %d' % tp.max_workers)
print('chunksize: %d' % tp.queue.maxsize)

# produce and send some data to the pool
print('First round:')
tp.feed([1, 2, 3, 4, 5])

print('Second round:')
tp.feed([6, 7, 8, 9, 0])

print('Large jobs:')
tp.feed(large_iterable(15))
# close the queue
tp.close()

print('Using callback (callback function will be threadsafe)')
tp = ThreadPool(my_worker, result_callback=callback)
tp.feed([1, 2, 3])
tp.close()