def get_session(name, password):
proxy = getip.getIP("")
url, yundama_obj, cid, session = do_login(name, password, proxy)
if url != '':
rs_cont = session.get(url, headers=headers, proxies=proxy)
login_info = rs_cont.text
u_pattern = r'"uniqueid":"(.*)",'
m = re.search(u_pattern, login_info)
if m and m.group(1):
# check if account is valid
check_url = 'http://weibo.com/2671109275/about'
resp = session.get(check_url, headers=headers, proxies=proxy)
if is_403(resp.text):
other.error('account {} has been forbidden'.format(name))
LoginInfoOper.freeze_account(name, 0)
return None
other.info(
'Login successful! The login account is {}'.format(name))
Cookies.store_cookies(name, session.cookies.get_dict(),
proxy['http'])
return session
other.error('login failed for {}'.format(name))
return None
def get_session(name, password, need_verify):
url, yundama_obj, cid, session = do_login(name, password, need_verify)
# 打码出错处理
while url == 'pinerror' and yundama_obj is not None:
yundama_obj.report_error(cid)
url, yundama_obj, cid, session = do_login(name, password, need_verify)
if url != '':
rs_cont = session.get(url, headers=headers)
login_info = rs_cont.text
u_pattern = r'"uniqueid":"(.*)",'
m = re.search(u_pattern, login_info)
if m and m.group(1):
# 任意验证一个页面看能否访问，使用这个方法验证比较依赖外部条件，但是没找到更好的方式(有的情况下，
# 账号存在问题，但是可以访问自己的主页，所以通过自己的主页验证账号是否正常不恰当)
check_url = 'http://weibo.com/p/1005051764222885/info?mod=pedit_more'
resp = session.get(check_url, headers=headers)
# 通过实验，目前发现未经过手机验证的账号是救不回来了...
if is_403(resp.text):
other.error('账号{}已被冻结'.format(name))
crawler.warning('账号{}已经被冻结'.format(name))
freeze_account(name, 0)
return None
other.info('本次登陆账号为:{}'.format(name))
Cookies.store_cookies(name, session.cookies.get_dict())
return session
other.error('本次账号{}登陆失败'.format(name))
return None
def get_session(name, password):
proxy = getip.getIP("")
url, yundama_obj, cid, session = do_login(name, password, proxy)
if url != '':
rs_cont = session.get(url, headers=headers, proxies=proxy)
login_info = rs_cont.text
u_pattern = r'"uniqueid":"(.*)",'
m = re.search(u_pattern, login_info)
if m and m.group(1):
# check if account is valid
check_url = 'http://weibo.com/2671109275/about'
resp = session.get(check_url, headers=headers, proxies=proxy)
if is_403(resp.text):
other.error('account {} has been forbidden'.format(name))
LoginInfoOper.freeze_account(name, 0)
return None
other.info('Login successful! The login account is {}'.format(name))
Cookies.store_cookies(name, session.cookies.get_dict(), proxy['http'])
return session
other.error('login failed for {}'.format(name))
return None
def login(self):
user_name = input('请输入登陆名: ')
password = input('请输入密码: ')
#判断登录方式
if re.match(r'\d{4}-\d{7}|1[34578]\d{9}', user_name):
print('当前登录方式:手机登陆')
post_url = 'https://www.zhihu.com/login/phone_num'
post_data = {
'phone_num': user_name,
'password': password,
'_xsrf': self.getXsrf()
}
else:
print('当前登录方式:邮箱登陆')
post_url = 'https://www.zhihu.com/login/email'
post_data = {
'email': user_name,
'password': password,
'_xsrf': self.getXsrf()
}
try:
login_page = self.session.post(post_url,
headers=headers,
data=post_data)
print(login_page.status)
print('直接登陆成功!')
except:
print('直接登陆失败!')
post_data['captcha'] = self.getCaptcha()
login_page = self.session.post(post_url,
headers=headers,
data=post_data)
Cookies.store_cookies(user_name, self.session.cookies.get_dict())
login_code = eval(login_page.text)
print(login_code['msg'])
def get_session(name, password):
url, yundama_obj, cid, session = do_login(name, password)
if url != '':
rs_cont = session.get(url, headers=headers)
login_info = rs_cont.text
u_pattern = r'"uniqueid":"(.*)",'
m = re.search(u_pattern, login_info)
if m and m.group(1):
# 访问微博官方账号看是否正常
check_url = 'http://weibo.com/2671109275/about'
resp = session.get(check_url, headers=headers)
# 通过实验，目前发现未经过手机验证的账号是救不回来了...
if is_403(resp.text):
other.error('账号{}已被冻结'.format(name))
freeze_account(name, 0)
return None
other.info('本次登陆账号为:{}'.format(name))
Cookies.store_cookies(name, session.cookies.get_dict())
return session
other.error('本次账号{}登陆失败'.format(name))
return None
def send_personal_message(target_uid, adver_message, user_verify=True, need_login=True):
"""
:param url: url to be crawled
:param user_verify: if it's ajax url, the value is False, else True
:param need_login: if the url is need to login, the value is True, else False
:return: return '' if exception happens or status_code != 200
"""
crawler.info('the send_personal_message uid is {uid}'.format(uid=str(target_uid)))
count = 0
while count < max_retries:
if need_login:
name_cookies = Cookies.fetch_cookies()
print(name_cookies)
# check adver_timers
if int(name_cookies[3]) >= int(adver_timers):
continue
if name_cookies is None:
crawler.warning('no cookies in cookies pool, please find out the reason')
send_email()
os.kill(os.getppid(), signal.SIGTERM)
try:
if need_login:
resp = requests.post('http://api.weibo.com/webim/2/direct_messages/new.json?source='+str(name_cookies[2]),
data={'text': adver_message, 'uid':str(target_uid)},
cookies=name_cookies[1], headers=personal_message_headers)
if "error" in resp.text:
crawler.warning('account {} has been banned, resp.text is: {}'.format(name_cookies[0], resp.text))
freeze_account(name_cookies[0], 0)
Cookies.delete_cookies(name_cookies[0])
continue
else:
# update adver_times
Cookies.store_cookies(name_cookies[0], name_cookies[1], name_cookies[2], 1)
return None
# if "$CONFIG['islogin'] = '******'" in resp.text:
# crawler.warning('account {} has been banned'.format(name_cookies[0]))
# freeze_account(name_cookies[0], 0)
# Cookies.delete_cookies(name_cookies[0])
# continue
# # else:
# # resp = requests.get(url, headers=headers, timeout=time_out, verify=False)
#
# page = resp.text
#
# if page:
# page = page.encode('utf-8', 'ignore').decode('utf-8')
# else:
# continue
#
# # slow down to aviod being banned
# time.sleep(interal)
#
# if user_verify:
# if is_banned(resp.url) or is_403(page):
# crawler.warning('account {} has been banned'.format(name_cookies[0]))
# freeze_account(name_cookies[0], 0)
# Cookies.delete_cookies(name_cookies[0])
# count += 1
# continue
#
# if 'verifybmobile' in resp.url:
# crawler.warning('account {} has been locked，you should use your phone to unlock it'.
# format(name_cookies[0]))
#
# freeze_account(name_cookies[0], -1)
# Cookies.delete_cookies(name_cookies[0])
# continue
#
# if not is_complete(page):
# count += 1
# continue
#
# if is_404(page):
# crawler.warning('send_personal_message{uid} seems to be 404'.format(uid=str(target_uid)))
# return ''
except (requests.exceptions.ReadTimeout, requests.exceptions.ConnectionError, AttributeError) as e:
crawler.warning('excepitons happens when send_personal_message {}，specific infos are {}'.format(target_uid, e))
count += 1
time.sleep(excp_interal)
else:
# Urls.store_crawl_url(url, 1)
# return page
return None
crawler.warning('max tries for {}，check the target_uid in redis db2'.format(target_uid))
# Urls.store_crawl_url(url, 0)
return ''
def get_session(name, password):
session = requests.Session()
su = get_encodename(name)
sever_data = get_server_data(su, session)
servertime = sever_data["servertime"]
nonce = sever_data['nonce']
rsakv = sever_data["rsakv"]
pubkey = sever_data["pubkey"]
sp = get_password(password, servertime, nonce, pubkey)
# 提交的数据可以根据抓包获得
data = {
'encoding': 'UTF-8',
'entry': 'weibo',
'from': '',
'gateway': '1',
'nonce': nonce,
'pagerefer': "",
'prelt': 67,
'pwencode': 'rsa2',
"returntype": "META",
'rsakv': rsakv,
'savestate': '7',
'servertime': servertime,
'service': 'miniblog',
'sp': sp,
'sr': '1920*1080',
'su': su,
'useticket': '1',
'vsnf': '1',
'url': 'http://weibo.com/ajaxlogin.php?framelogin=1&callback=parent.sinaSSOController.feedBackUrlCallBack'
}
post_url = 'http://login.sina.com.cn/sso/login.php?client=ssologin.js(v1.4.18)'
url = get_redirect(data, post_url, session)
if url != '':
rs_cont = session.get(url, headers=headers)
login_info = rs_cont.text
u_pattern = r'"uniqueid":"(.*)",'
m = re.search(u_pattern, login_info)
if m:
if m.group(1):
# 任意验证一个页面看能否访问，使用这个方法验证比较依赖外部条件，但是没找到更好的方式(有的情况下，
# 账号存在问题，但是可以访问自己的主页，所以通过自己的主页验证账号是否正常不恰当)
check_url = 'http://weibo.com/p/1005051764222885/info?mod=pedit_more'
resp = session.get(check_url, headers=headers)
if is_403(resp.text):
other.error('账号{}已被冻结'.format(name))
crawler.warning('账号{}已经被冻结'.format(name))
freeze_account(name)
return None
other.info('本次登陆账号为:{}'.format(name))
Cookies.store_cookies(name, session.cookies.get_dict())
return session
else:
other.error('本次账号{}登陆失败'.format(name))
return None
else:
other.error('本次账号{}登陆失败'.format(name))
return None
else:
other.error('本次账号{}登陆失败'.format(name))
return None
def test_store_and_fetch_cookies(self):
assert Cookies.fetch_cookies() is None
Cookies.store_cookies(FAKE_STR, FAKE_STR)
assert Cookies.fetch_cookies() is not None
def test_store_and_fetch_cookies(self):
assert Cookies.fetch_cookies() is None
Cookies.store_cookies(FAKE_STR, FAKE_STR)
assert Cookies.fetch_cookies() is not None