Python 3中的多线程文件下载类

今天在网上看到一个多线程文件下载类，觉得编码写的不错，但是是python2的，所以改造了一下，不废话，上代码,说明：我的代码都会自己测试通过，所以可以放心使用

我的博客：http://www.paulwangblog.com/

其中有一个小细节，那就是从HTTP包头中抓出文件大小，方法如下：
urlHandler = urllib.request.urlopen( url )#返回一个response对象
headers = urlHandler.info()#这里面就是头信息了

length = int(headers.get(‘Content-Length’))#用这个方法马上就能抓出你要的东西，网上没什么资料，找了1个多小时后放弃，最后看了源码才找到这个方法

'''
Created on 2011-11-10

@author: PaulWang

Description:
'''

'''It is a multi-thread downloading tool

    It was developed follow axel.
        Author: volans
        E-mail: volansw [at] gmail.com
'''

import sys
import os
import time
import urllib.request
from threading import Thread

#===============================================================================
# def download(url, output, blocks=6, proxies=local_proxies)
# output:输出文件的全路径，不带路径帽默认在本文件夹中生成
# blocks:分几块，开几个线程
# proxies:代理地址
#===============================================================================

local_proxies = {'http': 'http://131.139.58.200:8080'}#代理地址

class AxelPython(Thread, urllib.request.FancyURLopener):
    '''Multi-thread downloading class.

        run() is a vitural method of Thread.
    '''
    def __init__(self, threadname, url, filename, ranges=0, proxies={}):
        Thread.__init__(self, name=threadname)
        urllib.request.FancyURLopener.__init__(self, proxies)
        self.name = threadname
        self.url = url
        self.filename = filename
        self.ranges = ranges
        self.downloaded = 0

    def run(self):
        '''vertual function in Thread'''
        try:
            self.downloaded = os.path.getsize( self.filename )
        except OSError:
            #print 'never downloaded'
            self.downloaded = 0

        # rebuild start poind
        self.startpoint = self.ranges[0] + self.downloaded

        # This part is completed
        if self.startpoint >= self.ranges[1]:
            print ('Part %s has been downloaded over.' % self.filename)
            return

        self.oneTimeSize = 16384 #16kByte/time
        print ('task %s will download from %d to %d' % (self.name, self.startpoint, self.ranges[1]))

        self.addheader("Range", "bytes=%d-%d" % (self.startpoint, self.ranges[1]))

        self.urlhandle = self.open( self.url )

        data = self.urlhandle.read( self.oneTimeSize )
        while data:
            filehandle = open( self.filename, 'ab+' )
            filehandle.write( data )
            filehandle.close()

            self.downloaded += len( data )
            #print "%s" % (self.name)
            #progress = u'\r...'

            data = self.urlhandle.read( self.oneTimeSize )

def GetUrlFileSize(url, proxies={}):
    urlHandler = urllib.request.urlopen( url )
    headers = urlHandler.info()

    length = int(headers.get('Content-Length'))
    print('Content-Length is %d' % length)
    return length

def SpliteBlocks(totalsize, blocknumber):
    blocksize = totalsize/blocknumber
    ranges = []
    for i in range(0, blocknumber-1):
        ranges.append((i*blocksize, i*blocksize +blocksize - 1))
    ranges.append(( blocksize*(blocknumber-1), totalsize -1 ))

    return ranges
def islive(tasks):
    for task in tasks:
        if task.isAlive():
            return True
    return False

def download(url, output, blocks=6, proxies=local_proxies):
    ''' paxel
    '''
    size = GetUrlFileSize( url, proxies )
    ranges = SpliteBlocks( size, blocks )

    threadname = [ "thread_%d" % i for i in range(0, blocks) ]
    filename = [ "tmpfile_%d" % i for i in range(0, blocks) ]

    tasks = []
    for i in range(0,blocks):
        task = AxelPython( threadname[i], url, filename[i], ranges[i] )
        task.setDaemon( True )
        task.start()
        tasks.append( task )

    time.sleep( 2 )
    while islive(tasks):
        downloaded = sum( [task.downloaded for task in tasks] )
        process = downloaded/float(size)*100
        show = '\rFilesize:%d Downloaded:%d Completed:%.2f%%' % (size, downloaded, process)
        sys.stdout.write(show)
        sys.stdout.flush()
        time.sleep( 0.5 )

    filehandle = open( output, 'wb+' )
    for i in filename:
        f = open( i, 'rb' )
        filehandle.write( f.read() )
        f.close()
        try:
            os.remove(i)
            pass
        except:
            pass

    filehandle.close()

if __name__ == '__main__':
    url = "http://www.pygtk.org/dist/pygtk2-tut.pdf"
    output = 'pygtk2.pdf'
    download( url, output, blocks=1, proxies={} )
1