Learning about Queue module in python (how to run it)

Question

Was recently introduced to the queue design in regards to ability to defer processing as well as implementing a "FIFO" etc.

Looked through the documentation in attempt to get a sample queue going to understand how to implement it in my own design / program. But I'm having issues with just running this code:

import queue

def worker():
    while True:
        item = q.get()
        do_work(item)
        q.task_done()

def main():

    q = queue.Queue(maxsize=0)
    for i in range(num_worker_threads):
         t = Thread(target=worker)
         t.daemon = True
         t.start()

    for item in source():
        q.put(item)

    q.join()       # block until all tasks are done

main()

Question: Would like someone to explain what the for loops are doing, I get an error just running the code so I have to be missing something.

Problem Error that occurs: NameError: global name 'num_worker_threads' is not defined

Thank you from a -Python Novice-

Answer 1

The for loop is launching a number of worker threads to perform the function defined by "worker". Here is working code that should run on your system in python 2.7.

import Queue
import threading

# input queue to be processed by many threads
q_in = Queue.Queue(maxsize=0)

# output queue to be processed by one thread
q_out = Queue.Queue(maxsize=0)

# number of worker threads to complete the processing
num_worker_threads = 10

# process that each worker thread will execute until the Queue is empty
def worker():
    while True:
        # get item from queue, do work on it, let queue know processing is done for one item
        item = q_in.get()
        q_out.put(do_work(item))
        q_in.task_done()

# squares a number and returns the number and its square
def do_work(item):
    return (item,item*item)

# another queued thread we will use to print output
def printer():
    while True:
        # get an item processed by worker threads and print the result. Let queue know item has been processed
        item = q_out.get()
        print "%d squared is : %d" % item
        q_out.task_done()

# launch all of our queued processes
def main():
    # Launches a number of worker threads to perform operations using the queue of inputs
    for i in range(num_worker_threads):
         t = threading.Thread(target=worker)
         t.daemon = True
         t.start()

    # launches a single "printer" thread to output the result (makes things neater)
    t = threading.Thread(target=printer)
    t.daemon = True
    t.start()

    # put items on the input queue (numbers to be squared)
    for item in range(10):
        q_in.put(item)

    # wait for two queues to be emptied (and workers to close)   
    q_in.join()       # block until all tasks are done
    q_out.join()

    print "Processing Complete"

main()

Python 3 version per @handle

import queue 
import threading

# input queue to be processed by many threads
q_in = queue.Queue(maxsize=0) 

# output queue to be processed by one thread
q_out = queue.Queue(maxsize=0) 

# number of worker threads to complete the processing
num_worker_threads = 10

# process that each worker thread will execute until the Queue is empty
def worker():
    while True:
        # get item from queue, do work on it, let queue know processing is done for one item
        item = q_in.get()
        q_out.put(do_work(item))
        q_in.task_done()

# squares a number and returns the number and its square
def do_work(item):
    return (item,item*item)

# another queued thread we will use to print output
def printer():
    while True:
        # get an item processed by worker threads and print the result. Let queue know item has been processed
        item = q_out.get()
        print("{0[0]} squared is : {0[1]}".format(item) )
        q_out.task_done()

# launch all of our queued processes
def main():
    # Launches a number of worker threads to perform operations using the queue of inputs
    for i in range(num_worker_threads):
         t = threading.Thread(target=worker)
         t.daemon = True
         t.start()

    # launches a single "printer" thread to output the result (makes things neater)
    t = threading.Thread(target=printer)
    t.daemon = True
    t.start()

    # put items on the input queue (numbers to be squared)
    for item in range(10):
        q_in.put(item)

    # wait for two queues to be emptied (and workers to close)   
    q_in.join()       # block until all tasks are done
    q_out.join()

    print( "Processing Complete" )

main()

Answer 2

You can think of the number of worker threads as the number of bank tellers at a bank. So people (your items) stand in line (your queue) to be processed by a bank teller (your worker thread). Queues are actually an easy and well understood mechanism to manage complexities in threads.

I have adjusted your code a bit to show how it works.

import queue
import time
from threading import Thread

def do_work(item):
    print("processing", item)

def source():
    item = 1
    while True:
        print("starting", item)
        yield item
        time.sleep(0.2)
        item += 1

def worker():
    while True:
        item = q.get()
        do_work(item)
        q.task_done()

q = queue.Queue(maxsize=0)
def main():
    for i in range(2):
        t = Thread(target=worker)
        t.daemon = True
        t.start()

    for item in source():
        q.put(item)

    q.join()       # block until all tasks are done

main()