1. 程式人生 > >Python3 urllib庫和requests庫

Python3 urllib庫和requests庫

1. Python3 使用urllib庫請求網路

1.1 基於urllib庫的GET請求

請求百度首頁www.baidu.com ,不新增請求頭資訊:

 1 import urllib.requests
 2 
 3 
 4 def get_page():
 5   url = 'http://www.baidu.com/'
 6   res = urllib.request.urlopen(url=url)
 7   page_source = res.read().decode('utf-8')
 8   print(page_source)
 9 
10 
11 if __name__ == '__main__':
12   get_page()

輸出顯示百度首頁的原始碼。但是有的網站進行了反爬蟲設定,上述程式碼可能會返回一個40X之類的響應碼,因為該網站識別出了是爬蟲在訪問網站,這時需要偽裝一下爬蟲,讓爬蟲模擬使用者行為,給爬蟲設定headers(User-Agent)屬性,模擬瀏覽器請求網站。

1.2 使用User-Agent偽裝後請求網站

由於urllib.request.urlopen() 函式不接受headers引數,所以需要構建一個urllib.request.Request物件來實現請求頭的設定:

 1 import urllib.request
 2 
 3 
 4 def get_page():
 5   url = 'http://www.baidu.com'
 6   headers = {
 7     'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.146 Safari/537.36'
 8   }
 9   request = urllib.request.Request(url=url, headers=headers)
10   res = urllib.request.urlopen(request)
11   page_source = res.read().decode('utf-8')
12   print(page_source)
13 
14 
15 if __name__ == '__main__':
16   get_page()

新增headers引數,來模擬瀏覽器的行為。

1.3 基於urllib庫的POST請求,並用Cookie保持會話

登陸ChinaUnix論壇,獲取首頁原始碼,然後訪問一個文章。首先不使用Cookie看一下什麼效果:

 1 import urllib.request
 2 import urllib.parse
 3 
 4 
 5 def get_page():
 6   url = 'http://bbs.chinaunix.net/member.php?mod=logging&action=login&loginsubmit=yes&loginhash=LcN2z'
 7   headers = {
 8     'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.146 Safari/537.36'
 9   }
10   data = {
11     'username': 'StrivePy',
12     'password': 'XXX'
13   }
14   postdata = urllib.parse.urlencode(data).encode('utf-8')
15   req = urllib.request.Request(url=url, data=postdata, headers=headers)
16   res = urllib.request.urlopen(req)
17   page_source = res.read().decode('gbk')
18   print(page_source)
19 
20   url1 = 'http://bbs.chinaunix.net/thread-4263876-1-1.html'
21   res1 = urllib.request.urlopen(url=url1)
22   page_source1 = res1.read().decode('gbk')
23   print(page_source1)
24 
25 
26 if __name__ == '__main__':
27   get_page()

 

搜尋原始碼中是否能看見使用者名稱StrivePy,發現登陸成功,但是再請求其它文章時,顯示為遊客狀態,會話狀態沒有保持。現在使用Cookie看一下效果:

 1 import urllib.request
 2 import urllib.parse
 3 import http.cookiejar
 4 
 5 
 6 def get_page():
 7   url = 'http://bbs.chinaunix.net/member.php?mod=logging&action=login&loginsubmit=yes&loginhash=LcN2z'
 8   headers = {
 9     'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.146 Safari/537.36'
10   }
11   data = {
12     'username': 'StrivePy',
13     'password': 'XXX'
14   }
15   postdata = urllib.parse.urlencode(data).encode('utf-8')
16   req = urllib.request.Request(url=url, data=postdata, headers=headers)
17   # 建立CookieJar物件
18   cjar = http.cookiejar.CookieJar()
19   # 以CookieJar物件為引數建立Cookie
20   cookie = urllib.request.HTTPCookieProcessor(cjar)
21   # 以Cookie物件為引數建立Opener物件
22   opener = urllib.request.build_opener(cookie)
23   # 將Opener安裝位全域性,覆蓋urlopen函式,也可以臨時使用opener.open()函式
24   urllib.request.install_opener(opener)
25   res = urllib.request.urlopen(req)
26   page_source = res.read().decode('gbk')
27   print(page_source)
28 
29   url1 = 'http://bbs.chinaunix.net/thread-4263876-1-1.html'
30   res1 = urllib.request.urlopen(url=url1)
31   page_source1 = res1.read().decode('gbk')
32   print(page_source1)
33 
34 
35 if __name__ == '__main__':
36   get_page()

 

結果顯示登陸成功後,再訪問其它文章時,顯示為登陸狀態。若要將Cookie儲存為檔案待下次使用,可以使用MozillaCookieJar物件將Cookie儲存為檔案。

 1 import urllib.request
 2 import urllib.parse
 3 import http.cookiejar
 4 
 5 
 6 def get_page():
 7     url = 'http://bbs.chinaunix.net/member.php?mod=logging&action=login&loginsubmit=yes&loginhash=LcN2z'
 8     headers = {
 9         'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.146 Safari/537.36'
10     }
11     data = {
12         'username': 'StrivePy',
13         'password': 'XXX'
14     }
15     postdata = urllib.parse.urlencode(data).encode('utf-8')
16     req = urllib.request.Request(url=url, data=postdata, headers=headers)
17     filename = 'cookies.txt'
18     # 建立CookieJar物件
19     cjar = http.cookiejar.MozillaCookieJar(filename)
20     # 以CookieJar物件為引數建立Cookie
21     cookie = urllib.request.HTTPCookieProcessor(cjar)
22     # 以Cookie物件為引數建立Opener物件
23     opener = urllib.request.build_opener(cookie)
24     # 臨時使用opener來請求
25     opener.open(req)
26     # 將cookie儲存為檔案
27     cjar.save(ignore_discard=True, ignore_expires=True)

 

會在當前工作目錄生成一個名為cookies.txtcookie檔案,下次就可以不用登陸(如果cookie沒有失效的話)直接讀取這個檔案來實現免登入訪問。例如不進行登陸直接訪問其中一篇文章(沒登陸也可以訪問,主要是看擡頭是不是登陸狀態):

 1 import http.cookiejar
 2 
 3 
 4 def get_page():
 5     url1 = 'http://bbs.chinaunix.net/thread-4263876-1-1.html'
 6     filename = 'cookies.txt'
 7     cjar = http.cookiejar.MozillaCookieJar(filename)
 8     cjar.load(ignore_discard=True, ignore_expires=True)
 9     cookie = urllib.request.HTTPCookieProcessor(cjar)
10     opener = urllib.request.build_opener(cookie)
11     res1 = opener.open(url1)
12     page_source1 = res1.read().decode('gbk')
13     print(page_source1)
14 
15 
16 if __name__ == '__main__':
17     get_page()

 

結果顯示是以登陸狀態在檢視這篇文章。

1.4 基於urllib庫使用代理請求

使用代理可以有效規避爬蟲被封。

 1 import urllib.request
 2 
 3 
 4 def proxy_test():
 5     url = 'http://myip.kkcha.com/'
 6     headers = {
 7         'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.146 Safari/537.36'
 8     }
 9     request = urllib.request.Request(url=url, headers=headers)
10     proxy = {
11         'http': '180.137.232.101:53281'
12     }
13     # 建立代理Handler物件
14     proxy_handler = urllib.request.ProxyHandler(proxy)
15     # 以Handler物件為引數建立Opener物件
16     opener = urllib.request.build_opener(proxy_handler)
17     # 將Opener安裝為全域性
18     urllib.request.install_opener(opener)
19     response = urllib.request.urlopen(request)
20     page_source = response.read().decode('utf-8')
21     print(page_source)
22 
23 
24 if __name__ == '__main__':
25     proxy_test()

 

抓取到的頁面應該顯示代理IP,不知道什麼原因,有時候能正常顯示,有時候跳轉到有道詞典廣告頁!!!問題有待更進一步研究

2. Python3 使用requsets庫訪問網路

2.1 基於requests庫的GET請求

GET方式請求http://httpbin.org測試網站。

 1 import requests
 2 
 3 
 4 def request_test():
 5     url = 'http://httpbin.org/get'
 6     response = requests.get(url)
 7     print(type(response.text), response.text)
 8     print(type(response.content), response.content)
 9 
10 
11 if __name__ == '__main__':
12     request_test()

 

直接得到響應體。

1 <class 'str'> {"args":{},"headers":{"Accept":"*/*","Accept-Encoding":"gzip, deflate","Connection":"close","Host":"httpbin.org","User-Agent":"python-requests/2.18.4"},"origin":"121.61.132.191","url":"http://httpbin.org/get"}
2 
3 <class 'bytes'> b'{"args":{},"headers":{"Accept":"*/*","Accept-Encoding":"gzip, deflate","Connection":"close","Host":"httpbin.org","User-Agent":"python-requests/2.18.4"},"origin":"121.61.132.191","url":"http://httpbin.org/get"}\n

GET方法中傳遞引數的三種方式:

  • 將字典形式的引數用urllib.parse.urlencode()函式編碼成url引數:

     

    1 http://httpbin.org/key1=value1&key2=value2
  • 直接在urllib.request.get()函式中使用params引數:
    1 import requests
    2 
    3 if __name__ == '__main__':
    4     payload = {
    5         'key1': 'value1',
    6         'key2': 'value2'
    7     }
    8     response = requests.get('http://httpbin.org/get', params=payload)
    9     print(response.url)        

     

    1 http://httpbin.org/key1=value1&key2=value2
  • url直接包含引數:
    1 http://httpbin.org/get?key2=value2&key1=value1

     

2.2 基於requests庫的POST請求,並用session保持會話

登陸ChinaUnix論壇,獲取首頁原始碼,然後訪問一個文章。首先不使用Session看一下什麼效果:

 1 import requests
 3 
 4 
 5 def get_page():
 6   url = 'http://bbs.chinaunix.net/member.php?mod=logging&action=login&loginsubmit=yes&loginhash=LcN2z'
 7   headers = {
 8     'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.146 Safari/537.36'
 9   }
10   data = {
11     'username': 'StrivePy',
12     'password': 'XXX'
13   }
14   response = requests.post(url=url, data=data, headers=headers)
15   page_source = response.text
16   print(response.status_code)
17   print(page_source)
18 
19   url1 = 'http://bbs.chinaunix.net/thread-4263876-1-1.html'
20   response1 = requests.get(url=url1, headers=headers)
21   page_source1 = response1.text
22   print(response1.status_code)
23   print(page_source1)
24 
25 
26 if __name__ == '__main__':
27   get_page()

結果顯示訪問其它文章時為遊客模式。接下來用session來維持會話看一下效果:

 1 import requests
 2 
 3 
 4 def get_page():
 5   url = 'http://bbs.chinaunix.net/member.php?mod=logging&action=login&loginsubmit=yes&loginhash=LcN2z'
 6   headers = {
 7     'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.146 Safari/537.36'
 8   }
 9   data = {
10     'username': 'StrivePy',
11     'password': 'XXX'
12   }
13   session = requests.session()
14   response = session.post(url=url, data=data, headers=headers)
15   page_source = response.text
16   print(response.status_code)
17   print(page_source)
18 
19   url1 = 'http://bbs.chinaunix.net/thread-4263876-1-1.html'
20   response1 = session.get(url=url1, headers=headers)
21   page_source1 = response1.text
22   print(response1.status_code)
23   print(page_source1)
24 
25 
26 if __name__ == '__main__':
27   get_page()

 

結果顯示訪問其它文章時,顯示為登陸狀態,會話保持住了。使用session的效果類似於urllib庫臨時使用opener或者將opener安裝為全域性的效果。

2.3 基於requests庫使用代理請求

在requests庫中使用代理:

 1 import requests
 2 
 3 
 4 def proxy_test():
 5     url = 'http://myip.kkcha.com/'
 6     headers = {
 7         'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.146 Safari/537.36'
 8     }
 9     proxy = {
10         'https': '61.135.217.7: 80'
11     }
12     response = requests.get(url=url, headers=headers, proxies=proxy)
13     print(response.text)
14 
15 
16 if __name__ == '__main__':
17     proxy_test()