单线程实现多个定时器
NewTimer.py
#!/usr/bin/env python
from heapq import *
from threading import Timer
import threading
import uuid
import time
import datetime
import sys
import math
global TimerStamp
global TimerTimes
class CancelFail(Exception):
pass
class Slot(object):
def __init__(self, period=0, interval=1, function=None, args=[], kwargs={}):
self.period = period
self.pc = 0
self.interval = interval
self.fire = 0
self.id = uuid.uuid1()
self.function = function
self.args = args
self.kwargs = kwargs
#system resolution millisecond
class NewTimer(object):
#set enough time make thread sleep, when NewTimer empty set enoug time, too
#make sure sum of your timer call back function execute time shorter than resolution
#todo use a worker thread to operate timer call back function
def __init__(self, resolution=1000):
global TimerStamp
TimerStamp = int(time.time() * 1000)
self.nofire = sys.maxint #next fire time interval
self.firestamp = self.nofire + TimerStamp
self.resolution = resolution# 1s
self.lock = threading.RLock()
self.wait = dict()
self.ready = dict()
self._start()
""" private operate ready list """
def _addToReadyList(self, slot, firestamp):
box = dict( [ (slot.id, slot)])
if not self.ready.has_key( firestamp ):
self.ready.update( [(firestamp, box)] )
else:
boxs = self.ready.get(firestamp)
boxs.update( box )
def _delFromReadyList(self, slot):
boxs = self.ready.get(slot.fire)
try:
box = boxs.pop(slot.id)
if not boxs:
self.ready.pop(slot.fire)
except (AttributeError, KeyError):
raise CancelFail
""" inside """
def _start(self):
global TimerStamp
try:
self.firestamp = sorted( self.ready.keys() )[0]
stamp = float((TimerStamp + self.firestamp - int(time.time()*1000)))/1000
except IndexError:
self.firestamp = self.nofire
stamp = self.nofire
try:
self.timer.cancel()
except AttributeError:
pass
self.timer = Timer( stamp, self.hander)
self.timer.start()
def hander(self, *args, **kwargs):
""" find time arrive slot, do it function """
self.lock.acquire()
try:
boxs = self.ready.pop( self.firestamp )
slots = boxs.values()
except KeyError:
slots = []
for slot in slots:
if slot.period:
slot.pc += 1
if slot.pc != slot.period:
slot.fire = slot.interval + slot.fire
self._addToReadyList(slot, slot.fire)
elif slot.period == -1:
slot.fire = slot.interval + slot.fire
self._addToReadyList(slot, slot.fire)
""" """
self._start()
self.lock.release()
for slot in slots:
try:
slot.function(slot.args, slot.kwargs)
except Exception:
print "slot id %s, timer function fail" % slot.id
""" operate new timer manager itself """
def stop(self):
self.timer.cancel()
""" new timer manager """
def add(self, period=0, interval=1, function=None, args=[], kwargs={}):
"""
period: one time = 0, times = >0, always = -1
interval: timer fire relative TimerReference
function: when timer fire, call back function
args,kwargs: callback function args
"""
interval = int(interval) * self.resolution#seconds
if interval interval = self.resolution
slot = Slot( period, interval, function, *args, **kwargs )
box = dict([(slot.id, slot)])
self.wait.update(box)
return slot
def remove(self, slot):
if isinstance(slot, Slot):
self.cancel(slot)
try:
self.wait.pop(slot.id)
except KeyError:
print "wait dict not has the cancel timer"
""" timer api """
def reset(self, slot):
if isinstance(slot, Slot):
self.cancel(slot)
slot.pc = 0
self.start(slot)
def start(self, slot):
def NewTimerStamp(timebase, resolution):
nowoffset = int(time.time() * 1000) - timebase
if nowoffset % resolution currentstamp = nowoffset / resolution
else:
currentstamp = (nowoffset + resolution - 1) / resolution
return currentstamp * 1000
global TimerStamp
if isinstance(slot, Slot):
firestamp = slot.interval + NewTimerStamp(TimerStamp, self.resolution)
slot.fire = firestamp
self.lock.acquire()
self._addToReadyList(slot, firestamp)
if self.firestamp > slot.fire:
self._start()
self.lock.release()
def cancel(self, slot):
if isinstance(slot, Slot):
try:
self.lock.acquire()
self._delFromReadyList(slot)
self._start()
self.lock.release()
except CancelFail:
self.lock.release()
def hello( *args, **kargs):
print args[0], datetime.datetime.now()
if __name__ == "__main__":
print "start test timer", datetime.datetime.now()
nt = NewTimer(500)
t0 = nt.add( -1, 5, hello, [0])
t1 = nt.add( 4, 7, hello, [1])
t2 = nt.add( 1, 3, hello, [2])#
t3 = nt.add( 1, 4, hello, [3])#
t4 = nt.add( 4, 5, hello, [4])
t5 = nt.add( 12, 5, hello, [5])#
t6 = nt.add( 9, 7, hello, [6])
t7 = nt.add( 1, 8, hello, [7])#
t8 = nt.add( 40, 1, hello, [8])
nt.start( t0 )
nt.start( t1 )
nt.start( t2 )#
nt.start( t3 )#
nt.start( t4 )
nt.start( t5 )#
nt.start( t6 )
nt.start( t7 )#
nt.start( t8 )
nt.cancel(t2)
nt.cancel(t3)
nt.remove(t5)
nt.remove(t3)
time.sleep(3)
nt.start(t2)
nt.cancel(t8)
time.sleep(300)
nt.stop()
print "finish test timer", datetime.datetime.now()

Python is an interpreted language, but it also includes the compilation process. 1) Python code is first compiled into bytecode. 2) Bytecode is interpreted and executed by Python virtual machine. 3) This hybrid mechanism makes Python both flexible and efficient, but not as fast as a fully compiled language.

Useaforloopwheniteratingoverasequenceorforaspecificnumberoftimes;useawhileloopwhencontinuinguntilaconditionismet.Forloopsareidealforknownsequences,whilewhileloopssuitsituationswithundeterminediterations.

Pythonloopscanleadtoerrorslikeinfiniteloops,modifyinglistsduringiteration,off-by-oneerrors,zero-indexingissues,andnestedloopinefficiencies.Toavoidthese:1)Use'i

Forloopsareadvantageousforknowniterationsandsequences,offeringsimplicityandreadability;whileloopsareidealfordynamicconditionsandunknowniterations,providingcontrolovertermination.1)Forloopsareperfectforiteratingoverlists,tuples,orstrings,directlyacces

Pythonusesahybridmodelofcompilationandinterpretation:1)ThePythoninterpretercompilessourcecodeintoplatform-independentbytecode.2)ThePythonVirtualMachine(PVM)thenexecutesthisbytecode,balancingeaseofusewithperformance.

Pythonisbothinterpretedandcompiled.1)It'scompiledtobytecodeforportabilityacrossplatforms.2)Thebytecodeistheninterpreted,allowingfordynamictypingandrapiddevelopment,thoughitmaybeslowerthanfullycompiledlanguages.

Forloopsareidealwhenyouknowthenumberofiterationsinadvance,whilewhileloopsarebetterforsituationswhereyouneedtoloopuntilaconditionismet.Forloopsaremoreefficientandreadable,suitableforiteratingoversequences,whereaswhileloopsoffermorecontrolandareusefulf

Forloopsareusedwhenthenumberofiterationsisknowninadvance,whilewhileloopsareusedwhentheiterationsdependonacondition.1)Forloopsareidealforiteratingoversequenceslikelistsorarrays.2)Whileloopsaresuitableforscenarioswheretheloopcontinuesuntilaspecificcond


Hot AI Tools

Undresser.AI Undress
AI-powered app for creating realistic nude photos

AI Clothes Remover
Online AI tool for removing clothes from photos.

Undress AI Tool
Undress images for free

Clothoff.io
AI clothes remover

Video Face Swap
Swap faces in any video effortlessly with our completely free AI face swap tool!

Hot Article

Hot Tools

SublimeText3 Linux new version
SublimeText3 Linux latest version

Safe Exam Browser
Safe Exam Browser is a secure browser environment for taking online exams securely. This software turns any computer into a secure workstation. It controls access to any utility and prevents students from using unauthorized resources.

VSCode Windows 64-bit Download
A free and powerful IDE editor launched by Microsoft

PhpStorm Mac version
The latest (2018.2.1) professional PHP integrated development tool

MantisBT
Mantis is an easy-to-deploy web-based defect tracking tool designed to aid in product defect tracking. It requires PHP, MySQL and a web server. Check out our demo and hosting services.
