#首先我们需要导入 requests 库
import requests
# 请求的url
url = "https://top.chinaz.com/gongsitop/index_500top.html"
# 设置请求头信息
headers = {
    "user-agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/95.0.4638.69 Safari/537.36"
}
# 使用reqeusts模快发起 GET 请求
response = requests.get(url, headers=headers)
# 获取请求的返回结果
html = response.text
# 导入 re 模快
import re
# 存储内容
message = []
# 总共16个页面的数据
for page in range(16):
    # 组装url
    if page == 0:
        url = "https://top.chinaz.com/gongsitop/index_500top.html"
    else:
        url = "https://top.chinaz.com/gongsitop/index_500top_{}.html".format(page + 1)
    # 使用reqeusts模快发起 GET 请求
    response = requests.get(url, headers=headers)
    html = response.text
    # 使用 findall 函数来获取数据
    # 公司名
    company = re.findall('<a.*?target="_blank">(.+?)</a></h3>', html)
    #注册资本
    registered=re.findall('注册资本：</span>(.*?)</p>',html)
    pageOne = list(zip(company,registered))
    # 合并列表
    message.extend(pageOne)
# 导入python中的内置模块csv
import csv
with open("content.csv", "w") as f:
    w = csv.writer(f)
    w.writerows(message)
!cat content.csv
import pandas as pd

# 读取数据
df = pd.read_csv("content.csv", names=["company", "registered"])
df.head()
df.info()
# 在jupyter中直接展示图像
%matplotlib inline
import matplotlib.pyplot as plt

# 用黑体显示中文
plt.rcParams['font.sans-serif'] = ['SimHei']  

#
df['registered']=df['registered'].astype(str)
df['registered'] = df['registered'].str.extract('(\d+(?:\.\d+)?)',expand=False).astype(float)

#降序
df_sorted=df.sort_values(by='registered',ascending=False)

#提取前20位
top_20=df_sorted.head(20)

#柱状图
plt.bar(top_20['company'],top_20['registered'])
plt.xticks(rotation=45)
plt.xlabel('公司名字')
plt.ylabel('注册资本')

#
plt.show