|
发表于 2015-8-26 15:56:22
|
显示全部楼层
import urllib.request
import os
import random
def url_open(url):
data=None
#'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/31.0.1650.63 Safari/537.36'#
headers = {'User-Agent':'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/31.0.1650.63 Safari/537.36'}
req = urllib.request.Request(url,data,headers)
#proxies = ['117.177.243.43:8086']
#proxy = random.choice(proxies)
#'124.93.222.95:8080'
#proxy_support = urllib.request.ProxyHandler({"http":proxy})
#opener = urllib.request.build_opener(proxy_support)
#urllib.request.install_opener(opener)
response = urllib.request.urlopen(req)
html = response.read()
return html
def get_page(url):
html = url_open(url).decode('utf-8')
a = html.find('current-comment-page') + 23
b = html.find(']',a)
return html[a:b]
def find_imgs(url):
html = url_open(url).decode('utf-8')
img_addrs = []
a = html.find('img src=')
while a != -1:
b = html.find('.jpg',a,a+255)
if b != -1:
img_addrs.append(html[a+9:b+4])
else:
b = a+9
a = html.find('img src',b)
return img_addrs
def save_imgs(folder,img_addrs):
for each in img_addrs:
filename = each.split('/')[-1]
with open(filename,'wb') as f:
img =url_open(each)
f.write(img)
def download_mm(floder='煎蛋网爬虫',pages=10):
os.mkdir(floder)
os.chdir(floder)
url ="http://jandan.net/ooxx/"
page_num = int(get_page(url))
for i in range(pages):
page_num -= i
page_url = url + 'page-' + str(page_num) +'#comments'
img_addrs = find_imgs(page_url)
save_imgs(floder,img_addrs)
if __name__ == '__main__':
download_mm()
|
|