一、在這里selenium的作用
（1）模擬的登錄。
（2）獲取登錄成功之后的cookies
代碼
def start_login(self):
chrome_options = Options()
# 禁止圖片加載,禁止推送通知
prefs = {
"profile.default_content_setting_values": {
"images": 2
}, "profile.default_content_setting_values.notifications": 2
}
chrome_options.add_experimental_option("prefs", prefs)
if chrome_args().get("headless_flag") == "1":
chrome_options.add_argument(chrome_args().get("headless"))
chrome_options.add_argument(chrome_args().get("nogpu"))
chrome_options.add_argument(chrome_args().get("noinfobars"))
chrome_options.add_argument(chrome_args().get("max_windows"))
chrome_options.add_argument(self.Proxy_server)
driver = webdriver.Chrome(chrome_options=chrome_options)
try:
get_logger().info("start login.....")
try:
# login info
self.login_name = self.loginInfo.get("login_id")
self.password = aes_cbc_decrypt(self.loginInfo.get("login_pwd"))
except Exception:
get_logger().error("cant get login info,here are detals".format(traceback.format_exc()))
wait = WebDriverWait(driver, 30)
print(u"start login in")
driver.get(self.login_url)
try:
login_id = wait.until(
eccd.presence_of_element_located(
(By.XPATH, self.Id_xpath))
)
login_id.send_keys(self.login_name)
login_id.send_keys(Keys.ENTER)
password = wait.until(
eccd.presence_of_element_located(
(By.XPATH, self.pwd_xpath))
)
password.send_keys(self.password)
submit = wait.until(
eccd.presence_of_element_located(
(By.XPATH, self.login_btn_xpath))
)
submit.click()
# login signal
#判斷是否顯示 右上角是否顯示用戶名
login_ok = wait.until(
eccd.presence_of_element_located(
(By.XPATH, self.login_ok_xpath))
)
try:
#判斷登錄成功
if login_ok:
get_logger().info("get user name successful:"+u"{}".format(login_ok.text))
try:
get_weibo_info=driver.find_element_by_xpath(self.forward_home_page)
except:
driver.get(self.first_page)
forward_home_page_ok = wait.until(
eccd.presence_of_element_located(
(By.XPATH, self.forward_home_page))
)
forward_home_page_ok.click()
time.sleep(5)
self.first_page=driver.current_url
get_logger().info("get homepage successful,url is {}".format(driver.current_url))
cookies = driver.get_cookies() # 導出cookie
get_logger().info("get cookies")
get_fansnum_ok= wait.until(
eccd.presence_of_element_located(
(By.XPATH, self.follow_xpath))
)
self.get_followfanshome(driver)
get_logger().info("get_followfanshome ok")
self.home_page_source=driver.page_source
driver.close() # 關閉chrome
#獲取主頁的
# 如果需要保存cookies
self.write_cookie(cookies)
get_logger().info("get cookies,login ok")
return cookies
else:
raise RuntimeError('login failed')
except:
get_logger().error("login failed")
return None
except:
if driver is not None:
driver.close()
except Exception as e:
if driver is not None:
driver.close()
get_logger().error("dbs operation error，here are details:{}".format(traceback.format_exc()))
通過使用cookies = driver.get_cookies() #我們獲取了cookie那么我們怎用呢。
二、requests如何使用獲取到的cookies
1.創建一個Session對象
req = requests.Session() # 構建Session
2.轉換上面的cookies對象
for cookie in cookies:
req.cookies.set(cookie['name'], cookie['value']) # 轉換cookies
3.開始訪問該網站的需求頁面
data = req.get(url, headers=self.headers).text #獲取文本網頁 jsondata = req.get(url, headers=self.headers).json() #獲取json型網頁
好了就是這么簡單，另外補充一個經常使用的庫fake-useragent，可以獲取隨機的useragent，但是其中谷歌的useragent版本比較低，尤其在訪問知乎的時候，會提示瀏覽器版本低的信息，這一點需要注意。
另外擁有一個headers是對爬蟲起碼的尊重，所以有個構建好的headers是非常必要的。我就經常用下面的例：
from fake_useragent import UserAgent as UA
import random
headers = {
'Accept': 'text/html, application/xhtml+xml, image/jxr, */*',
'Accept - Encoding': 'gzip, deflate',
'Accept-Language': 'zh-Hans-CN, zh-Hans; q=0.5',
'Connection': 'Keep-Alive',
'User-Agent': UA().random #獲取隨機的useragent
}