Reintroduce threadpool and filesize packages

This commit is contained in:
Hugo Posnic 2017-11-18 16:16:59 +01:00
parent 25836634af
commit 278ef269cf
7 changed files with 405 additions and 1 deletions

View file

@ -0,0 +1,280 @@
'''
ThreadPool Implementation
@author: Morten Holdflod Moeller - morten@holdflod.dk
@license: LGPL v3
'''
from __future__ import with_statement
from threading import Thread, RLock
from time import sleep
from Queue import Queue, Empty
import logging
import sys
class NullHandler(logging.Handler):
def emit(self, record):
pass
h = sys.stderr
logging.getLogger('threadpool').addHandler(h)
logging.getLogger('threadpool.worker').addHandler(h)
class ThreadPoolMixIn:
"""Mix-in class to handle each request in a new thread from the ThreadPool."""
def __init__(self, threadpool=None):
if (threadpool == None):
threadpool = ThreadPool()
self.__private_threadpool = True
else:
self.__private_threadpool = False
self.__threadpool = threadpool
def process_request_thread(self, request, client_address):
"""Same as in BaseServer but as a thread.
In addition, exception handling is done here.
"""
try:
self.finish_request(request, client_address)
self.close_request(request)
except:
self.handle_error(request, client_address) #IGNORE:W0702
self.close_request(request)
def process_request(self, request, client_address):
self.__threadpool.add_job(self.process_request_thread, [request, client_address])
def shutdown(self):
if (self.__private_threadpool): self.__threadpool.shutdown()
class AddJobException(Exception):
'''
Exceptoion raised when a Job could not be added
to the queue
'''
def __init__(self, msg):
Exception.__init__(self, msg)
class ThreadPool:
'''
The class implementing the ThreadPool.
Instantiate and add jobs using add_job(func, args_list)
'''
class Job: #IGNORE:R0903
'''
Class encapsulating a job to be handled
by ThreadPool workers
'''
def __init__(self, function, args, return_callback=None):
self.callable = function
self.arguments = args
self.return_callback = return_callback
def execute(self):
'''
Called to execute the function
'''
try:
return_value = self.callable(*self.arguments) #IGNORE:W0142
except Exception, excep: #IGNORE:W0703
logger = logging.getLogger("threadpool.worker")
logger.warning("A job in the ThreadPool raised an exception: " + excep)
#else do nothing cause we don't know what to do...
return
try:
if (self.return_callback != None):
self.return_callback(return_value)
except Exception, _: #IGNORE:W0703 everything could go wrong...
logger = logging.getLogger('threadpool')
logger.warning('Error while delivering return value to callback function')
class Worker(Thread):
'''
A worker thread handling jobs in the thread pool
job queue
'''
def __init__(self, pool):
Thread.__init__(self)
if (not isinstance(pool, ThreadPool)):
raise TypeError("pool is not a ThreadPool instance")
self.pool = pool
self.alive = True
self.start()
def run(self):
'''
The workers main-loop getting jobs from queue
and executing them
'''
while self.alive:
#print self.pool.__active_worker_count, self.pool.__worker_count
job = self.pool.get_job()
if (job != None):
self.pool.worker_active()
job.execute()
self.pool.worker_inactive()
else:
self.alive = False
self.pool.punch_out()
def __init__(self, max_workers = 5, kill_workers_after = 3):
if (not isinstance(max_workers, int)):
raise TypeError("max_workers is not an int")
if (max_workers < 1):
raise ValueError('max_workers must be >= 1')
if (not isinstance(kill_workers_after, int)):
raise TypeError("kill_workers_after is not an int")
self.__max_workers = max_workers
self.__kill_workers_after = kill_workers_after
# This Queue is assumed Thread Safe
self.__jobs = Queue()
self.__worker_count_lock = RLock()
self.__worker_count = 0
self.__active_worker_count = 0
self.__shutting_down = False
logger = logging.getLogger('threadpool')
logger.info('started')
def shutdown(self, wait_for_workers_period = 1, clean_shutdown_reties = 5):
if (not isinstance(clean_shutdown_reties, int)):
raise TypeError("clean_shutdown_reties is not an int")
if (not clean_shutdown_reties >= 0):
raise ValueError('clean_shutdown_reties must be >= 0')
if (not isinstance(wait_for_workers_period, int)):
raise TypeError("wait_for_workers_period is not an int")
if (not wait_for_workers_period >= 0):
raise ValueError('wait_for_workers_period must be >= 0')
logger = logging.getLogger("threadpool")
logger.info("shutting down")
with self.__worker_count_lock:
self.__shutting_down = True
self.__max_workers = 0
self.__kill_workers_after = 0
retries_left = clean_shutdown_reties
while (retries_left > 0):
with self.__worker_count_lock:
logger.info("waiting for workers to shut down (%i), %i workers left"%(retries_left, self.__worker_count))
if (self.__worker_count > 0):
retries_left -= 1
else:
retries_left = 0
sleep(wait_for_workers_period)
with self.__worker_count_lock:
if (self.__worker_count > 0):
logger.warning("shutdown stopped waiting. Still %i active workers"%self.__worker_count)
clean_shutdown = False
else:
clean_shutdown = True
logger.info("shutdown complete")
return clean_shutdown
def punch_out(self):
'''
Called by worker to update worker count
when the worker is shutting down
'''
with self.__worker_count_lock:
self.__worker_count -= 1
def __new_worker(self):
'''
Adding a new worker thread to the thread pool
'''
with self.__worker_count_lock:
ThreadPool.Worker(self)
self.__worker_count += 1
def worker_active(self):
with self.__worker_count_lock:
self.__active_worker_count = self.__active_worker_count + 1
def worker_inactive(self):
with self.__worker_count_lock:
self.__active_worker_count = self.__active_worker_count - 1
def add_job(self, function, args = None, return_callback=None):
'''
Put new job into queue
'''
if (not callable(function)):
raise TypeError("function is not a callable")
if (not ( args == None or isinstance(args, list))):
raise TypeError("args is not a list")
if (not (return_callback == None or callable(return_callback))):
raise TypeError("return_callback is not a callable")
if (args == None):
args = []
job = ThreadPool.Job(function, args, return_callback)
with self.__worker_count_lock:
if (self.__shutting_down):
raise AddJobException("ThreadPool is shutting down")
try:
start_new_worker = False
if (self.__worker_count < self.__max_workers):
if (self.__active_worker_count == self.__worker_count):
start_new_worker = True
self.__jobs.put(job)
if (start_new_worker):
self.__new_worker()
except Exception:
raise AddJobException("Could not add job")
def get_job(self):
'''
Retrieve next job from queue
workers die (and should) when
returning None
'''
job = None
try:
if (self.__kill_workers_after < 0):
job = self.__jobs.get(True)
elif (self.__kill_workers_after == 0):
job = self.__jobs.get(False)
else:
job = self.__jobs.get(True, self.__kill_workers_after)
except Empty:
job = None
return job

View file

@ -0,0 +1 @@
from ThreadPool import ThreadPool, ThreadPoolMixIn

View file

@ -0,0 +1,71 @@
Metadata-Version: 1.0
Name: hurry.filesize
Version: 0.9
Summary: A simple Python library for human readable file sizes (or anything sized in bytes).
Home-page: UNKNOWN
Author: Martijn Faassen, Startifact
Author-email: faassen@startifact.com
License: ZPL 2.1
Description: hurry.filesize
==============
hurry.filesize a simple Python library that can take a number of bytes and
returns a human-readable string with the size in it, in kilobytes (K),
megabytes (M), etc.
The default system it uses is "traditional", where multipliers of 1024
increase the unit size::
>>> from hurry.filesize import size
>>> size(1024)
'1K'
An alternative, slightly more verbose system::
>>> from hurry.filesize import alternative
>>> size(1, system=alternative)
'1 byte'
>>> size(10, system=alternative)
'10 bytes'
>>> size(1024, system=alternative)
'1 KB'
A verbose system::
>>> from hurry.filesize import verbose
>>> size(10, system=verbose)
'10 bytes'
>>> size(1024, system=verbose)
'1 kilobyte'
>>> size(2000, system=verbose)
'1 kilobyte'
>>> size(3000, system=verbose)
'2 kilobytes'
>>> size(1024 * 1024, system=verbose)
'1 megabyte'
>>> size(1024 * 1024 * 3, system=verbose)
'3 megabytes'
You can also use the SI system, where multipliers of 1000 increase the unit
size::
>>> from hurry.filesize import si
>>> size(1000, system=si)
'1K'
Changes
=======
0.9 (2009-03-11)
----------------
* Initial public release.
Download
========
Keywords: file size bytes
Platform: UNKNOWN
Classifier: Programming Language :: Python
Classifier: Topic :: Software Development :: Libraries :: Python Modules

View file

@ -0,0 +1,50 @@
hurry.filesize
==============
hurry.filesize a simple Python library that can take a number of bytes and
returns a human-readable string with the size in it, in kilobytes (K),
megabytes (M), etc.
The default system it uses is "traditional", where multipliers of 1024
increase the unit size::
>>> from hurry.filesize import size
>>> size(1024)
'1K'
An alternative, slightly more verbose system::
>>> from hurry.filesize import alternative
>>> size(1, system=alternative)
'1 byte'
>>> size(10, system=alternative)
'10 bytes'
>>> size(1024, system=alternative)
'1 KB'
A verbose system::
>>> from hurry.filesize import verbose
>>> size(10, system=verbose)
'10 bytes'
>>> size(1024, system=verbose)
'1 kilobyte'
>>> size(2000, system=verbose)
'1 kilobyte'
>>> size(3000, system=verbose)
'2 kilobytes'
>>> size(1024 * 1024, system=verbose)
'1 megabyte'
>>> size(1024 * 1024 * 3, system=verbose)
'3 megabytes'
You can also use the SI system, where multipliers of 1000 increase the unit
size::
>>> from hurry.filesize import si
>>> size(1000, system=si)
'1K'
http://pypi.python.org/pypi/hurry.filesize

View file

@ -0,0 +1,2 @@
from filesize import size
from filesize import traditional, alternative, verbose, iec, si

View file

@ -19,7 +19,7 @@ from filesize import *
from imghdr import what as determinetype
from queue import Queue
from ThreadPool import ThreadPool, ThreadPoolMixIn
from ThreadPool import ThreadPool
from multiprocessing import cpu_count
from ui import Ui_trimage