获取中国大学排名前400名

在终端显示前400名大学

import requests
from bs4 import BeautifulSoup
import bs4
import re
def getHTMLtext(url):
    try:
        r = requests.get(url,timeout = 30)
        r.raise_for_status   #查看响应是否是200
        r.encoding = r.apparent_encoding
        return r.text
    except:
        return ''
def fillUnivList(ulist,html):
    soup = BeautifulSoup(html,'html.parser')
    for tr in soup.find('tbody').children:
        if isinstance(tr,bs4.element.Tag):   #判断是否是标签
            tds = tr('td')   #找出所有的td标签
            ulist.append([tds[0].string,tds[1].string,tds[3].string])
def printUnivList(ulist,num):
     tplt = "{0:^10}\t{1:{3}^10}\t{2:^10}"
     print(tplt.format("排名","学校名称","总分",chr(12288)))   #chr(12288)以中文字符填充空格
     for i in range(num):
        u = ulist[i]
        print(tplt.format(u[0],u[1],u[2],chr(12288)))
def main():
    ulist = []
    url = 'http://www.zuihaodaxue.cn/zuihaodaxuepaiming2019.html'
    html = getHTMLtext(url)
    fillUnivList(ulist,html)
    printUnivList(ulist,400)

main()

运行结果：

将前400名大学存入MySQL

创建存放数据的MySQL表：
create table university(
-> id int(5) not null,
-> name char(30) not null,
-> score float(5) not null);

import requests
from bs4 import BeautifulSoup
import bs4
import re
import pymysql


class university():
     def __init__(self,url,num):
          self.url = url
          self.ulist = []
          self.num = num
          self.db = pymysql.connect(host='127.0.0.1',port = 3306,user='root',password = 'forever1122',db='python')
     def getHTMLtext(self):
          try:
               r = requests.get(self.url,timeout = 30)
               r.raise_for_status       #查看响应是否是200
               r.encoding = r.apparent_encoding
               html = r.text
               return html
          except:
               return ''
     def fillUnivList(self):
          html = self.getHTMLtext()
          soup = BeautifulSoup(html,'html.parser')
          for tr in soup.find('tbody').children:
               if isinstance(tr,bs4.element.Tag):   #判断tr是否是标签
                    tds = tr('td')   #找出所有的td标签
                    self.ulist.append([tds[0].string,tds[1].string,tds[3].string]) #获取想要的排名，校名，总分
     def printUnivList(self):
          for i in range(self.num):
               u = self.ulist[i]
               id = u[0]    #排名
               name = u[1]    #校名
               score = u[2]     #总分
               sql = 'insert into university (id,name,score) values ({},"{}",{})'.format(id,name,score)
               self.write_to_mysql(sql)
          print("输入完成！！！！！")
     def write_to_mysql(self,sql):
          cursor = self.db.cursor()
          try:
               cursor.execute(sql)
               self.db.commit()
          except Exception as e:
               print(e)
               self.db.rollback()

if __name__=="__main__":
     un = university('http://www.zuihaodaxue.cn/zuihaodaxuepaiming2019.html',400)
     un.getHTMLtext()
     un.fillUnivList()
     un.printUnivList()

运行结果：