def update_headers(cls):
cls.logger.info('*********updating cookies*********')
_, headers, _ = parse_curl_str(cls.curl_str)
headers['Cookie'] = cls.get_cookie_str()
if headers['Cookie'] is None:
change_ip()
cls.update_headers()
else:
cls.headers = headers
def update_headers(cls):
cls.logger.info('*********updating cookies*********')
_, headers, _ = parse_curl_str(cls.curl_str)
headers['Cookie'] = cls.get_cookie_str()
if headers['Cookie'] is None:
change_ip()
cls.update_headers()
else:
cls.headers = headers
def update_headers(self, changeip=True):
if changeip:
change_ip()
r = get(self.base_url)
h = cookie_dict_from_cookie_str(r.headers.get('Set-Cookie'))
cookies_dict = cookie_dict_from_cookie_str(self.headers['Cookie'])
cookies_dict.update(h)
self.headers['Cookie'] = cookies_dict
self.logger.info('headers: %s', pformat(self.headers))
def handle_response(self, url, response):
self.logger.info('handle url: %s', url)
if not response:
return
if response.status_code == 200:
html = response.text
html_parser = KuaidailiHtmlParser(url, html)
ip_info_dict_yield = html_parser.parse()
self.bulk_update_to_mongo(ip_info_dict_yield)
elif response.status_code == 503:
change_ip()
self.urls.append(url) # retry
def handle_response(self, url, response):
"""handle_response 把代理ip的信息存储到mongodb中
:param url:
:param response: requests.models.Response
"""
self.logger.info('handle url: %s', url)
if not response:
return
if response.status_code == 200:
html = response.text
html_parser = XiciHtmlParser(url, html)
ip_info_dict_yield = html_parser.parse()
self.bulk_update_to_mongo(ip_info_dict_yield)
elif response.status_code == 503:
change_ip()
self.urls.append(url) # retry
def fetch_channel_json(self, channel_json_url):
time.sleep(random.randint(30, 60))
self.logger.info(channel_json_url)
res = get(channel_json_url, headers=self.headers)
# http://stackoverflow.com/questions/24027589/how-to-convert-raw-javascript-object-to-python-dictionary
html = res.text.strip()
o = ast.literal_eval(html)
if not o:
self.logger.info(pprint.pformat(html))
self.logger.info('fetch channel_json_url: %s failed',
channel_json_url)
change_ip()
return
nick_name = o['nick_name']
general_msg_list = o['general_msg_list']
article_list = ast.literal_eval(general_msg_list)['list']
article_dict_list = []
for article in article_list:
app_msg_ext_info = article['app_msg_ext_info']
comm_msg_info = article['comm_msg_info']
ori_create_time = comm_msg_info['datetime']
article_dict_list.append(
self._get_articel_info(app_msg_ext_info, nick_name,
ori_create_time))
if app_msg_ext_info['is_multi']:
for article_info in app_msg_ext_info[
'multi_app_msg_item_list']:
article_dict_list.append(
self._get_articel_info(article_info, nick_name,
ori_create_time))
article_dict_list = self.get_remove_too_old_days_article(
article_dict_list)
article_dict_list = self.get_remove_mongodb_already_has_article(
nick_name, article_dict_list)
for article_dict in article_dict_list:
article_dict['link'] = self.get_permanent_wechat_article_url(
article_dict['link'])
self.logger.info(pprint.pformat(article_dict_list))
self.save_article_dict_list(nick_name, article_dict_list)
def fetch_channel_json(self, channel_json_url):
time.sleep(random.randint(60, 120))
self.logger.info(channel_json_url)
res = get(channel_json_url, headers=self.headers)
# http://stackoverflow.com/questions/24027589/how-to-convert-raw-javascript-object-to-python-dictionary
html = res.text.strip()
o = ast.literal_eval(html)
if not o:
self.logger.debug(pprint.pformat(html))
self.logger.info(
'fetch channel_json_url: %s failed', channel_json_url
)
change_ip()
return
nick_name = o['nick_name']
general_msg_list = o['general_msg_list']
article_list = ast.literal_eval(general_msg_list)['list']
article_dict_list = []
for article in article_list:
app_msg_ext_info = article['app_msg_ext_info']
comm_msg_info = article['comm_msg_info']
ori_create_time = comm_msg_info['datetime']
article_dict_list.append(
self._get_articel_info(
app_msg_ext_info, nick_name, ori_create_time
)
)
if app_msg_ext_info['is_multi']:
for article_info in app_msg_ext_info['multi_app_msg_item_list']:
article_dict_list.append(
self._get_articel_info(
article_info, nick_name, ori_create_time
)
)
article_dict_list = self.get_remove_too_old_days_article(article_dict_list)
article_dict_list = self.get_remove_mongodb_already_has_article(nick_name, article_dict_list)
self.logger.info(pprint.pformat(article_dict_list))
self.save_article_dict_list(nick_name, article_dict_list)
def _wrapper(*args, **kwargs):
index = 0
while index < retries:
index += 1
try:
response = func(*args, **kwargs)
if response and (
LagouCrawler.is_block_html(response.text) or
LagouCrawler.is_check_html(response.text)
):
sleep_time = (sleep ** index + random.randint(1, 10))
if sleep_time > 300: # 5 mins
change_ip()
continue
else:
print('sleep for %ds' % sleep_time)
time.sleep(sleep_time)
continue
if response.status_code in (301, 302, 404, 500):
print('status_code', response.status_code)
break
elif response.status_code != 200:
print(response.status_code)
if changeip:
change_ip()
continue
else:
break
except Exception as e:
traceback.print_exc()
response = None
if isinstance(e, Timeout):
if sleep is not None:
time.sleep(sleep + random.randint(10, 15))
continue
elif isinstance(e, TooManyRedirects):
break
return response
