解决某网站css加密改进说明

前言

取名最新是因为确实目前是最新的，三个月后会去掉最新
找了很多博客代码都过时了，而且不是很好懂
以前是字体库，现在是css加密，大众点评的反爬机制确实强
本篇主要是对https://blog.csdn.net/qq_40235133/article/details/116588731这篇博客的补充说明
建议先食用这篇，大概懂了原理，出bug时再来看我这篇
本篇主要是对原博客三个报错的说明改进

代码

import requests
import re
from lxml import etree
import pandas as pd
import time


def crew(i):
    """
    爬取第i页的评论
    :return: 第i页评论的表格
    """

    # 首先获取原始网页数据
    headers = {
        'Cookie': "_lxsdk_cuid=1841c99b9d8c8-0f1e3463ba50be-26021f51-144000-1841c99b9d8c8; _lxsdk=1841c99b9d8c8-0f1e3463ba50be-26021f51-144000-1841c99b9d8c8; _hc.v=7a55c341-cdde-f085-91b9-46cf79b3e9c4.1666927149; fspop=test; cy=5; cye=nanjing; _lx_utm=utm_source%3DBaidu%26utm_medium%3Dorganic; Hm_lvt_602b80cf8079ae6591966cc70a3940e7=1666927149,1666927732,1666934687; WEBDFPID=w7955624006058uxy4x7y89u3076vxu58158w9xxv01979589v3v53uz-1982294689448-1666934689197KSIQKMGfd79fef3d01d5e9aadc18ccd4d0c95075294; dplet=5c64382712747088c3c4439a13928211; dper=d0a6f2f41b6117319faebcb359b0362c20578bb6b4bfbd0f0b7cd0ad09f94b97008071297a9de353aff98bcf2c9fe5906601a93859b5dd80d999ab192ccbba6831a3195d1ff78d000576a2c958289c81a729e2c0d86cc5d6a1a5f12ad1d395d3; ll=7fd06e815b796be3df069dec7836c3df; ua=dpuser_3122839151; ctu=60b777f0958d913f993ce809bf1d930ee786702ba9e11effc2704689466c1d67; s_ViewType=10; Hm_lpvt_602b80cf8079ae6591966cc70a3940e7=1666934731; _lxsdk_s=1841d0cc474-c48-686-a8b%7C%7C177",
        'Host': 'www.dianping.com',
        'Referer': f'https://www.dianping.com/shop/i2r0VzqmFBkNLQXC/review_all/p{i}',
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/106.0.0.0 Safari/537.36'
    }
    response = requests.get(f'https://www.dianping.com/shop/i2r0VzqmFBkNLQXC/review_all/p{i}', headers=headers)
    with open('01 网页数据_加密.html', mode='w', encoding='utf-8') as f:
        f.write(response.text)

    # 获取css文件请求svg内容
    css_url = 'http:' + re.findall(r'<link rel="stylesheet" type="text/css" href="(//s3plus.meituan.*?)">', response.text)[0]
    css_response = requests.get(css_url)
    with open('02 css样式.css', mode='w', encoding='utf-8') as f:
        f.write(css_response.text)

    # 获取svg映射表
    svg_url  ='http:'+ re.findall(r'svgmtsi\[class\^="muv"\].*?background-image: url\((.*?)\);',css_response.text)[0]
    svg_response = requests.get(svg_url)
    with open('03 svg映射表.svg',mode='w',encoding='utf-8') as f:
        f.write(svg_response.text)

    # 获取svg加密字典
    import parsel
    with open('03 svg映射表.svg',mode='r',encoding='utf-8') as f:
        svg_html = f.read()
    sel = parsel.Selector(svg_html)
    paths = sel.css("path")
    texts = sel.css('textPath')
    lines = []
    for path,textPath in zip(paths,texts):
        lines.append([int("".join(re.findall("M0 (.*?) H600",path.css('path::attr(d)').get()))),textPath.css('textPath::text').get()])

    with open('02 css样式.css',mode='r',encoding='utf-8') as f:
        css_text = f.read()
    class_map = re.findall(r'\.(muv\w+)\{background:-(\d+)\.0px -(\d+)\.0px;\}',css_text)
    class_map = [(cls_name,int(x),int(y)) for cls_name,x,y in class_map]
    d_map = {}
    #获取类名与汉字的对应关系
    for one_char in class_map:
        try:
            cls_name,x,y =one_char
            for line in lines:
                if line[0] < y:
                    pass
                else:
                    index = int(x/14)
                    char = line[1][index]
    #                 print(cls_name,char)
                    d_map[cls_name] = char
                    break
        except Exception as e:
            print(e)

    # 替换svg加密字体，还原评论
    with open('01 网页数据_加密.html',mode='r',encoding='utf-8') as f:
        html = f.read()
    for key,value in d_map.items():
        html = html.replace('<svgmtsi class="'+key+'"></svgmtsi>',value)
    with open('04 网页数据_解密.html',mode='w',encoding='utf-8') as f:
        f.write(html)

    # 去除空格，并存到dataframe中
    e = etree.HTML(html)
    pl = e.xpath("//div[@class='review-words Hide']/text()")
    dq_list =[]
    cc_df = pd.DataFrame()
    for p in pl:
        if p == '\n\t                    ' or p == '\n                            ':
            pass
        else:
            dq_list.append(p)
    for pp in dq_list:
        cc_df = cc_df.append({'评论':pp}, ignore_index=True)

    return cc_df

def add_workbook():
    with open("out.csv","ab") as f:
        f.write(open("new_out.csv","rb").read())
    return None

if __name__ == '__main__':
    # 设置开始页数和终止界面
    start_page = 2  # 默认第一遍从第2页开始
    end_page = 100  # 默认爬取200页


    df1 = crew(1)
    # 开始爬虫
    try:
        for i in range(start_page,end_page):
            df2 = crew(i)
            df1 = pd.concat([df1,df2],axis=0,ignore_index=True)
            time.sleep(10)
            print(f"第{i}页爬取成功")
        
        # 存入excel
        df1.reset_index(inplace=True)
        df1.to_excel('new_out.xlsx', index=False)

    except:
        print(f"当前页数{i}")
        print("cookie已过期，请重置，并更新开始页数")
    # print(1)

说明

cookie

• 这个cookie表示登录状态，可能十分钟就要换

• 点这个网站：https://www.dianping.com/shop/i2r0VzqmFBkNLQXC/review_all

• 用自己的号登陆

• 搜索自己要获取评论的店、记得要点到评论页里

• 右键、选择检查

• 刷新网页

• 按图中四步就能找到cookie，然后复制替换掉原来py脚本里的过期cookie
  

• cookie不更新就会报如下错

获取css文件那块
IndexError: list index out of range

一些说明

• 这两处的muv，只针对这个店，换一家店可能就变了

  • 根据自己要访问的店的加密方式改，比如这家店加密的svg是muv

  • 

  • 

  • 搞错加密方式就会报如下错

svg映射表那块
list index out of range

• 大众点评查找方式就两种，我这里写的是第一种，根据id定位行的，原博客是根据x坐标定位行的，也许会改，看情况用代码就行
  • 根据id定位，但用了原博客代码的就会报如下错

lines.append([int(text.css('text::attr(y)').get())
ERROR：int() argument must be a string, a bytes-like object or a number, not 'NoneType'

• 根据xy定位，但用了我的代码
  • 不会报错但生成的excel会有点问题
  • 
  • 解决办法
    • 用上面的替换下面这块
  • 
    代码

	texts = sel.css('text')
    lines = []
    for text in texts:
        lines.append([int(text.css('text::attr(y)').get()), text.css('text::text').get()])

有什么问题可以在评论区问，趁我对这个还有记忆的时候