中国供应商字体反爬

我也不知道该怎么介绍这个网站，想不出来开场白，那就直奔主题吧

1. 要爬取的网站和信息
   中国供应商
   某个商品的商家电话
2. 查看电话信息是否加密
   
   不出意外是加密的，还判定是字体加密，那么就是常规思路了
3. 获取加密字体并查看规律
   
   保存为woff格式后进行查看
   

文字内容都是一样的 只是顺序不一样，是动态的
4. 查找相同点，构造映射关系

根据一个字体文件的这些数据的md5值进行构造字典

font_dict = {
    '0072296251cdc1cf39f36bfe71030516':1,
    'ad2cb8eb87212068a1f5fcd890decfad':3,
    '13164316728f72504c55a2ffb0e0c495':5,
    '168b3dc2e5f487ac5d2921af154182e1':7,
    'b651cdcb5c2a4234204c3219fb9bb82f':8,
    'f1d930f3bf8e87dd52e3b92c5bf5fdd7':2,
    'a6a070db8e1d3d6af10e04b203323641':9,
    '5b46b9d4d383db9e425efcfd2dd53cd9':0,
    '7edd56fae59e9cabd2655f138f2c0925':'-',
    'accd825aab52df41a5bc6701ff2156a6':6,
    'bf4e2e8facb4a13551e6ee343f44c022':4,

}

5. 然后就是每次请求的时候再根据新的字体文件里面的文字数据的md5值和上步构造好的映射做比对，再进行替换就可以了

代码如下 写作不易，点个三连吧

import requests
import base64
import re
from fontTools.ttLib import TTFont
from lxml import etree
from hashlib import md5
url = 'https://www.china.cn/leddengdai/4255707828.html'
headers = {
'user-agent': 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/88.0.4324.96 Safari/537.36'
}
ret = requests.get(url=url,headers=headers)


with open('china.html','w',encoding='utf8') as f:
    f.write(ret.text)

with open('china.html','r',encoding='utf8') as f:
    ret = f.read()
ba64 = re.findall('base64,(.*?)\'\) forma',ret)[0]
b= base64.b64decode(ba64)
with open('2.woff','wb') as f:
    f.write(b)
font = TTFont('2.woff')
font.saveXML('2.xml')

num_list = [1,3,5,7,8,2,9,0,'-',4]
font_dict = {
    '0072296251cdc1cf39f36bfe71030516':1,
    'ad2cb8eb87212068a1f5fcd890decfad':3,
    '13164316728f72504c55a2ffb0e0c495':5,
    '168b3dc2e5f487ac5d2921af154182e1':7,
    'b651cdcb5c2a4234204c3219fb9bb82f':8,
    'f1d930f3bf8e87dd52e3b92c5bf5fdd7':2,
    'a6a070db8e1d3d6af10e04b203323641':9,
    '5b46b9d4d383db9e425efcfd2dd53cd9':0,
    '7edd56fae59e9cabd2655f138f2c0925':'-',
    'accd825aab52df41a5bc6701ff2156a6':6,
    'bf4e2e8facb4a13551e6ee343f44c022':4,

}
font = TTFont('2.woff')
cmap = font.getBestCmap()
print(cmap)
with open('2.xml','r',encoding='utf8') as f:
    ret = f.read()
ret = ret.replace('\n','').replace(' ','')

font_cmap = {}
for i in cmap:
    data = re.findall(f'<CharStringname="{cmap[i]}">(.*?)</CharString>', ret)[0]
    char_md5 = md5(data.encode('utf8')).hexdigest()
    for k in font_dict:
        if k==char_md5:
            font_cmap['&#x'+cmap[i][3:]+';']=font_dict[k]
print(font_cmap)

with open('china.html','r',encoding='utf8') as f:
    ret = f.read()
for i in font_cmap:
    ret = ret.replace(i,str(font_cmap[i]))
page_html = etree.HTML(ret)
phone = page_html.xpath('/html//div[5]/div[2]/div[3]/div[1]/div[3]/span[1]/text()')
numb = page_html.xpath('/html//div[5]/div[2]/div[3]/div[1]/div[3]/span[2]/text()')

print(phone,numb)