多线程下载脚本

版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。

本文链接： https://blog.csdn.net/qq_20539533/article/details/102581728

《疯狂pthon讲义》第15章的一个多线程下载脚本。

实现多线程下载功能。

原本代码：

#!usr/bin/env python3.6
#encoding:utf-8

'''
author:shadow
datetime:2019/10/14
explain:
测试用例
多线程
下载
'''
import threading
from urllib.request import *

class DownThread(threading.Thread):
    def __init__(self, path,start_pos, current_part_size, current_part):
        super().__init__()
        self.path = path
        self.start_pos = start_pos
        self.current_part_size = current_part_size
        self.current_part = current_part
        self.length = 0
    def run(self):
        req = Request(url = self.path,  method='GET')
        req.add_header('Accept', '*/*')
        req.add_header('Charset', 'UTF-8')
        req.add_header('Connection', 'Keep-Alive')
        #打开下载资源
        f = urlopen(req)
        for i in range(self.start_pos):
            f.read(1)
        while self.length<self.current_part_size:
            data = f.read(1024)
            if data is None or len(data)<=0:
                break
            self.current_part.write(data)
            self.length+=len(data)
        self.current_part.close()
        f.close()


class DownLoadUnit:
    def __init__(self, path, target_file, thread_num):
        self.path = path
        self.thread_num = thread_num
        self.target_file = target_file
        self.threads = []
    def download(self):
        req = Request(url=self.path,method='GET')
        req.add_header('Accept', '*/*')
        req.add_header('Charset', 'UTF-8')
        req.add_header('Connection', 'Keep-Alive')
        #打开下载资源
        f = urlopen(req)
        #获取要下载的文件大小
        self.file_size = int(dict(f.headers).get('Content-Length', 0))
        f.close()
        #计算每个线程要下载的资源大小
        current_part_size = self.file_size // self.thread_num + 1
        for i in range(self.thread_num):
            #计算每个线程下载的开始位置
            start_pos = i * current_part_size
            #每个线程都用一个以wb模式打开的文件进行下载
            t = open(self.target_file, 'wb')
            #定位该线程的下载位置
            t.seek(start_pos, 0)
            td = DownThread(self.path, start_pos, current_part_size, t)
            self.threads.append(td)
            td.start()

    #获取下载百分比
    def get_compete_rate(self):
        sum_size = 0
        for i in range(self.thread_num):
            sum_size+=self.threads[i].length
        return sum_size/self.file_size


if __name__ == '__main__':
    d = DownLoadUnit("http://www.crazyit.org/data/attachment/forum/201801/19/121212ituj1s9gj8g880jr.png","a.jpg",4)
    d.download()
    def show_process():
        print("已完成:%.2f%%"%d.get_compete_rate())
        global c
        if d.get_compete_rate()<1:
            c=threading.Timer(0.1,show_process)
            c.start()


    c = threading.Timer(0.1, show_process)
    c.start()

难点：实现断点下载

解决方法：增加一个配置文件。