import requests
import os,sys
import time
import re


def get_html(http_session,url):
    r = http_session.get(url)
    r.encoding = r.apparent_encoding
    html = r.text
    return html

def get_imgurl(html):
    pic_urls = re.findall('"pic_url":"(.*?)"', html, re.S)
    img_urls = []
    for one_pic_url in pic_urls:
        img_urls.append('http:' + one_pic_url)
        print('http:' + one_pic_url)
    return img_urls

def write_tofile(img_urls,keyword,page):
    for id,img_url in enumerate(img_urls):
        pic = requests.get(img_url)
        with open(keyword+'/{}_{}.jpg'.format(page,id),'wb') as f:
            f.write(pic.content)

def main(http_session,url,keyword,page):
    html = get_html(http_session,url)
    img_urls = get_imgurl(html)
    write_tofile(img_urls,keyword,page)

if __name__=='__main__':
############################### cookie 准备 ###########################################
    ori_cookie = "..." ###通过 浏览器F12，Network可找到
    newcookie = ori_cookie.split(";")
    cookie = {}
    for c in newcookie:
        new_c = c.split('=',1)
        cookie[new_c[0]]=new_c[1]
    http_session = requests.session()
    requests.utils.add_dict_to_cookiejar(http_session.cookies,cookie)

############################### 输入关键字，开始爬取######################################
    keyword = "电脑" ###爬取关键字图片
    Page = 5  ###爬取页数
    if not os.path.exists(keyword):
        os.mkdir(keyword)
    for p in range(Page):
        url = 'http://s.taobao.com/search?q=' + keyword + '&s=' + str(p*44)
        print(url)
        main(http_session,url,keyword,p)
        time.sleep(10)
######################################################################################