网页编程网

# -*- coding: utf-8 -*-
"""
第1页，就是0，方便合起来写。每页的url找其规律
有一部电影没有，用三目运算符处理
列表换个位置
"""
import requests
from bs4 import BeautifulSoup
import openpyxl
headers={
        'user-agent':'Mozilla/5.0 (Windows NT 6.1; ) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/82.0.4068.4 Safari/537.36'
        }
lists=[['编号','名称','推荐','评分','链接']]
for page in range(10):
    url='https://movie.douban.com/top250?start={}&filter='.format(page*25)
    print(url) 
    response=requests.get(url,headers=headers)
    bs=BeautifulSoup(response.text,'html.parser')#解析器html.parser
#    title=bs.find('span',class_='title')
#    有多个相同的，故用findAll
    grid_view=bs.find('ol',class_='grid_view')#所有ol中
    all_li=grid_view.findAll('li')
    for i in all_li:
        number=i.find('em').text
        title=i.find('span',class_='title').text
        tuijian=i.find('span',class_='inq')
        scorce=i.find('span',class_='rating_num').text
        url=i.find('a')['href']
        lists.append([number,title,tuijian.text if tuijian !=None else '',scorce,url])
'''       
for ii in lists:
    print(ii)
'''
wb=openpyxl.Workbook()
sheet=wb.active
sheet.title='我喜欢的电影'
for item in lists:
    sheet.append(item)
wb.save('films.xlsx')