def get_redirect(name, data, post_url, session, proxy):
logining_page = session.post(post_url,
data=data,
headers=headers,
proxies=proxy)
login_loop = logining_page.content.decode("GBK")
# if name or password is wrong, set the value to 2
if 'retcode=101' in login_loop:
crawler.error(
'invalid password for {}, please ensure your account and password'.
format(name))
LoginInfoOper.freeze_account(name, 2)
return ''
if 'retcode=2070' in login_loop:
crawler.error('invalid verification code')
return 'pinerror'
if 'retcode=4049' in login_loop:
crawler.warning('account {} need verification for login'.format(name))
return 'login_need_pincode'
if '正在登录' in login_loop or 'Signing in' in login_loop:
pa = r'location\.replace\([\'"](.*?)[\'"]\)'
return re.findall(pa, login_loop)[0]
else:
return ''
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):
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 test_login_oper(self):
infos = LoginInfoOper.get_login_info()
assert len(infos) == 0
db_session.execute("insert into {} ({}.name) values (".format(
login_info.name, login_info.name) + FAKE_ID + ")")
infos = LoginInfoOper.get_login_info()
assert len(infos) == 1
LoginInfoOper.freeze_account(FAKE_ID, 0)
infos = LoginInfoOper.get_login_info()
assert len(infos) == 0
def test_login_oper(self):
infos = LoginInfoOper.get_login_info()
assert len(infos) == 0
db_session.execute("insert into {} ({}.name) values (".format(login_info.name, login_info.name)
+ FAKE_ID + ")")
infos = LoginInfoOper.get_login_info()
assert len(infos) == 1
LoginInfoOper.freeze_account(FAKE_ID, 0)
infos = LoginInfoOper.get_login_info()
assert len(infos) == 0
def execute_login_task():
infos = LoginInfoOper.get_login_info()
# Clear all stacked login tasks before each time for login
Cookies.check_login_task()
crawler.info('The login task is starting...')
for info in infos:
app.send_task('tasks.login.login_task',
args=(info.name, info.password),
queue='login_queue',
routing_key='for_login')
time.sleep(10)
def get_redirect(name, data, post_url, session, proxy):
logining_page = session.post(post_url, data=data, headers=headers, proxies=proxy)
login_loop = logining_page.content.decode("GBK")
# if name or password is wrong, set the value to 2
if 'retcode=101' in login_loop:
crawler.error('invalid password for {}, please ensure your account and password'.format(name))
LoginInfoOper.freeze_account(name, 2)
return ''
if 'retcode=2070' in login_loop:
crawler.error('invalid verification code')
return 'pinerror'
if 'retcode=4049' in login_loop:
crawler.warning('account {} need verification for login'.format(name))
return 'login_need_pincode'
if '正在登录' in login_loop or 'Signing in' in login_loop:
pa = r'location\.replace\([\'"](.*?)[\'"]\)'
return re.findall(pa, login_loop)[0]
else:
return ''
def execute_login_task():
# 获取所有的需要登录的weibo账号信息
infos = LoginInfoOper.get_login_info()
# Clear all stacked login tasks before each time for login
Cookies.check_login_task()
crawler.info('The login task is starting...')
for info in infos:
# 对xx任务 发送参数args
# 让这个任务启动
# queue参数：表示通过这个队列来路由通知任务
# 路由的key由参数routing_key 指定
app.send_task('tasks.login.login_task',
args=(info.name, info.password),
queue='login_queue',
routing_key='for_login')
time.sleep(10)
def get_page(url, auth_level=2, is_ajax=False, need_proxy=False):
"""
:param url: url to crawl
:param auth_level: 0 stands for need nothing,1 stands for no login but need cookies,2 stands for need login.
:param is_ajax: whether the request is ajax
:param need_proxy: whether the request need a http/https proxy
:return: response text, when a exception is raised, return ''
"""
crawler.info('the crawling url is {url}'.format(url=url))
count = 0
while count < MAX_RETRIES:
if auth_level == 2:
name_cookies = Cookies.fetch_cookies()
if name_cookies is None:
crawler.warning(
'No cookie in cookies pool. Maybe all accounts are banned, or all cookies are expired'
)
send_email()
os.kill(os.getppid(), signal.SIGTERM)
# There is no difference between http and https address.
proxy = {
'http': name_cookies[2],
'https': name_cookies[2],
}
else:
proxy = getip.getIPWithoutLogin('')
# if proxy['http'] is None:
# crawler.warning('No available ip in ip pools. Using local ip instead.')
try:
if auth_level == 2:
resp = requests.get(url,
headers=headers,
cookies=name_cookies[1],
timeout=TIME_OUT,
verify=False,
proxies=proxy)
elif auth_level == 1:
resp = requests.get(url,
headers=headers,
cookies=COOKIES,
timeout=TIME_OUT,
verify=False,
proxies=proxy)
else:
resp = requests.get(url,
headers=headers,
timeout=TIME_OUT,
verify=False,
proxies=proxy)
except (requests.exceptions.ReadTimeout,
requests.exceptions.ConnectionError, AttributeError) as e:
crawler.warning(
'Excepitons are raised when crawling {}.Here are details:{}'.
format(url, e))
count += 1
time.sleep(EXCP_INTERAL)
continue
if resp.status_code == 414:
crawler.warning('This ip has been blocked by weibo system')
if not need_proxy:
send_email()
os.kill(os.getppid(), signal.SIGTERM)
if resp.text:
page = resp.text.encode('utf-8', 'ignore').decode('utf-8')
else:
count += 1
continue
if auth_level == 2:
# slow down to aviod being banned
time.sleep(INTERAL)
if is_banned(resp.url) or is_403(page):
crawler.warning('Account {} has been banned'.format(
name_cookies[0]))
LoginInfoOper.freeze_account(name_cookies[0], 0)
Cookies.delete_cookies(name_cookies[0])
count += 1
continue
if not is_ajax and not is_complete(page):
count += 1
continue
if is_404(page):
crawler.warning('{} seems to be 404'.format(url))
return ''
Urls.store_crawl_url(url, 1)
return page
Urls.store_crawl_url(url, 0)
return ''
def get_page(url, auth_level=2, is_ajax=False, need_proxy=False):
"""
:param url: url to crawl
:param auth_level: 0 stands for need nothing,1 stands for no login but need cookies,2 stands for need login.
:param is_ajax: whether the request is ajax
:param need_proxy: whether the request need a http/https proxy
:return: response text, when a exception is raised, return ''
"""
crawler.info('the crawling url is {url}'.format(url=url))
count = 0
while count < MAX_RETRIES:
if auth_level == 2:
name_cookies = Cookies.fetch_cookies()
if name_cookies is None:
crawler.warning('No cookie in cookies pool. Maybe all accounts are banned, or all cookies are expired')
send_email()
os.kill(os.getppid(), signal.SIGTERM)
# There is no difference between http and https address.
proxy = {'http': name_cookies[2], 'https': name_cookies[2], }
else:
proxy = getip.getIPWithoutLogin('')
# if proxy['http'] is None:
# crawler.warning('No available ip in ip pools. Using local ip instead.')
try:
if auth_level == 2:
resp = requests.get(url, headers=headers, cookies=name_cookies[1], timeout=TIME_OUT, verify=False, proxies=proxy)
elif auth_level == 1:
resp = requests.get(url, headers=headers, cookies=COOKIES, timeout=TIME_OUT, verify=False, proxies=proxy)
else:
resp = requests.get(url, headers=headers, timeout=TIME_OUT, verify=False, proxies=proxy)
except (requests.exceptions.ReadTimeout, requests.exceptions.ConnectionError, AttributeError) as e:
crawler.warning('Excepitons are raised when crawling {}.Here are details:{}'.format(url, e))
count += 1
time.sleep(EXCP_INTERAL)
continue
if resp.status_code == 414:
crawler.warning('This ip has been blocked by weibo system')
if not need_proxy:
send_email()
os.kill(os.getppid(), signal.SIGTERM)
if resp.text:
page = resp.text.encode('utf-8', 'ignore').decode('utf-8')
else:
count += 1
continue
if auth_level == 2:
# slow down to aviod being banned
time.sleep(INTERAL)
if is_banned(resp.url) or is_403(page):
crawler.warning('Account {} has been banned'.format(name_cookies[0]))
LoginInfoOper.freeze_account(name_cookies[0], 0)
Cookies.delete_cookies(name_cookies[0])
count += 1
continue
if not is_ajax and not is_complete(page):
count += 1
continue
if is_404(page):
crawler.warning('{} seems to be 404'.format(url))
return ''
Urls.store_crawl_url(url, 1)
return page
Urls.store_crawl_url(url, 0)
return ''
