import requests url = "https://www.bitpush.news/covid19/" 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"} response = requests.get(url, headers=headers) html = response.text from lxml import etree parse = etree.HTMLParser(encoding='utf-8')
doc = etree.HTML(html) country = doc.xpath('//div[@class="table_container"]//tbody/tr/td/span/text()') person = doc.xpath('//div[@class="table_container"]//tbody/tr/td[2]/text()') death = doc.xpath('//div[@class="table_container"]//tbody/tr/td[3]/text()') death = [x.replace(",", "") for x in death] message = list(zip(country, person, death)) message import csv with open("content.csv", "w") as f: w = csv.writer(f) w.writerows(message) import pandas as pd df = pd.read_csv("content.csv", names=["country", "person", "death"], encoding='gb2312') df.info() df1 = df.drop(0).head(10)

import matplotlib.pyplot as plt %matplotlib inline plt.rcParams['font.sans-serif'] = ['SimHei'] plt.rcParams['figure.figsize'] = (10, 5)
x = df1["country"].values y = df1["death"].values plt.bar(x, y) plt.xlabel("国家",fontsize=14) plt.ylabel("死亡人数",fontsize=14) plt.show()