python学习之旅-爬虫

# -*- coding: utf-8 -*-
"""
Spyder Editor

This is a temporary script file.
"""

import re
import urllib.request

def getHtml(url):
    page = urllib.request.urlopen(url)
    html = page.read()
    return html

def getImg(html):    
    html = html.decode('utf_8')
    reg = r'src="(.*?\.jpg)" width'
    imgre = re.compile(reg)
    imglist = imgre.findall(html)
    return imglist

html = getHtml('https://movie.douban.com/')
x = 0
for imgurl in getImg(html):
    urllib.request.urlretrieve(imgurl,'%s.jpg' % x)
    x += 1

print(getImg(html))