反反爬 解决CloudFlare 5秒盾

Posted 2023-04-24

参考技术A 当我们第一次访问使用 CloudFlare 加速的网站时，网站就会出现让我们等待 5 秒种的提示

方法一、使用浏览器模拟技术请求目标网站，例如：Selenium
方法二、一个专门为了绕过这个 CloudFlare 开发的 Python 库 cloudscraper

用上它就可以无感爬取使用了 CloudFlare 的网站，使用这个库非常简单

文档 https://pypi.org/project/cloudscraper/

spider.py

middlewares.py

python反反爬，爬取猫眼评分

python反反爬，爬取猫眼评分.
解决网站爬取时，内容类似：$#x12E0;样式,且每次字体文件变化。
下载FontCreator

.

用FontCreator打开base.woff.查看对应字体关系

初始化时将对应关系写入字典中。

 1 #!/usr/bin/env python
 2 # coding:utf-8
 3 # __author__ = "南楼"
 4 
 5 
 6 import requests
 7 import re
 8 import os
 9 
10 from fontTools.ttLib import TTFont
11 
12 #下载字体
13 class MaoYan(object):
14 
15     def __init__(self):
16         self.url = ‘http://maoyan.com/films/1198214‘
17         self.headers = {
18             "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/59.0.3071.115 Safari/537.36"
19         }
20         self.base_num = {}  # 编号—数字
21         self.base_obj = {}  # 编号—对象
22         # base.woff 为当前网站下载的一个字体
23         self.base_font_file = TTFont(‘./fonts/base.woff‘)
24         # 需要先下载字体编辑软件(FontCreator),以便查看对应关系
25         self.base_num["uniF3BA"] = "0"
26         self.base_num["uniF2A9"] = "1"
27         self.base_num["uniE6A5"] = "2"
28         self.base_num["uniF680"] = "3"
29         self.base_num["uniE69C"] = "4"
30         self.base_num["uniE710"] = "5"
31         self.base_num["uniE07D"] = "6"
32         self.base_num["uniE5A7"] = "7"
33         self.base_num["uniEC7A"] = "8"
34         self.base_num["uniE2A3"] = "9"
35 
36         for key in self.base_num:
37             self.base_obj[key] =self.base_font_file[‘glyf‘][key]
38 
39     def baseobj(self):
40         for key in self.base_num:
41 
42             self.base_obj[key] =self.base_font_file[‘glyf‘][key]  # 获得woff内编号对应的字体对象
43             return self.base_obj
44 
45     # 发送请求获得响应
46     def get_html(self, url):
47         response = requests.get(url, headers=self.headers)
48         return response.content
49 
50     def create_font(self, re_font_file):
51         # 列出已下载文件
52         file_list = os.listdir(‘./fonts‘)
53         # 判断是否已下载
54         if re_font_file not in file_list:
55 
56             print(‘不在字体库中, 下载:‘, re_font_file)
57             url = ‘http://vfile.meituan.net/colorstone/‘ + re_font_file
58             new_file = self.get_html(url)
59             with open(‘./fonts/‘ + re_font_file, ‘wb‘) as f:
60                 f.write(new_file)
61 
62         # 打开字体文件，创建 self.font_file属性
63         self.font_file = TTFont(‘./fonts/‘ + re_font_file)
64 
65     def get_num_from_font_file(self, re_star):
66 
67         newstar = re_star.upper().replace("&#X", "uni")
68         realnum = newstar.replace(";", "")
69         numlist = realnum.split(".")
70         # gly_list = self.font_file.getGlyphOrder() #uni列表[‘glyph00000‘, ‘x‘, ‘uniF680‘, ‘uniE2A3‘, ‘uniE710‘, ‘uniE69C‘, ‘uniEC7A‘, ‘uniF2A9‘, ‘uniE5A7‘, ‘uniE07D‘, ‘uniE6A5‘, ‘uniF3BA‘]
71         star_rating = []
72         for hax_num in numlist:
73             font_file_num = self.font_file[‘glyf‘][hax_num]
74             for key in self.baseobj():
75                 if font_file_num == self.base_obj[key]:
76                     star_rating.append(self.base_num[key])
77         # 星级评分待优化，暂不支持10.0，
78         star_rating = star_rating[0]+"."+star_rating[1]
79         return star_rating
80 
81     def start_crawl(self):
82         html = self.get_html(self.url).decode(‘utf-8‘)
83 
84         # 正则匹配字体文件
85         re_font_file = re.findall(r‘vfile\\.meituan\\.net\\/colorstone\\/(\\w+\\.woff)‘, html)[0]
86         self.create_font(re_font_file)
87         # 正则匹配星级评分
88         re_star_rating = re.findall(r‘<span class="index-left info-num ">\\s+<span class="stonefont">(.*?)</span>\\s+</span>‘, html)[0]
89         star_rating = self.get_num_from_font_file(re_star_rating)
90         print("星级评分：", star_rating)
91 
92 
93 if __name__ == ‘__main__‘:
94 
95     m = MaoYan()
96     m.start_crawl()