Skip to content

Instantly share code, notes, and snippets.

@bigeagle
Created August 1, 2014 13:59
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save bigeagle/40e6a3dd94051970af6b to your computer and use it in GitHub Desktop.
Save bigeagle/40e6a3dd94051970af6b to your computer and use it in GitHub Desktop.
#!/usr/bin/env python2
# -*- coding:utf-8 -*-
import requests
import time
from pyquery import PyQuery as pq
from urlparse import urlparse, parse_qs
backup_file = "backup.txt"
base_url = "http://3g.renren.com/"
status_list_url = base_url + "/status/getdoing.do?sid={}"
headers = {
"Host": "3g.renren.com",
"Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
"User-Agent": "Mozilla/5.0 (X11; Linux x86_64; rv:30.0) Gecko/20100101 Firefox/30.0",
"Referer": base_url,
}
login_info = {
"email": "",
"password": "",
}
loginPage = requests.get(base_url).content
dom = pq(loginPage)
login_action_url = dom(".sec > form").attr("action")
login_data = {pq(x).attr("name"): pq(x).attr("value") for x in dom(".sec > form input")}
for k, v in login_info.iteritems():
login_data[k] = v
r = requests.post(login_action_url, headers=headers, data=login_data)
cookies = r.history[0].cookies
queries = parse_qs(urlparse(r.url).query)
sid = queries['sid'][0]
last_page = False
while not last_page:
r = requests.get(status_list_url.format(sid))
# r = requests.get(s_url, headers=headers)
dom = pq(r.content)
if not dom(".list > div.l"):
last_page = True
else:
print dom(".list > div.l > span.gray").text()
for i, status_dom in enumerate(dom(".list > div:not(.l)").items()):
_links = status_dom.children("a")
status_del_url = _links.eq(len(_links)-1).attr("href")
t = status_dom.children(".time").remove().text()
status_dom.children("a").remove()
status_dom.children("em").remove()
print status_del_url
with open(backup_file, "a") as f:
f.write(t.encode("utf_8"))
f.write(",\t")
f.write(status_dom.text().encode("utf-8"))
f.write("\n")
r = requests.get(status_del_url)
dom = pq(r.content)
del_action_url = dom(".sform > form").attr("action")
post_data = {pq(x).attr("name"): pq(x).attr("value") for x in dom(".sform > form input[type='hidden']")}
requests.post(del_action_url, post_data)
time.sleep(10)
print "删完了"
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment