注：以下代码需要使用Selenium、BeautifulSoup、json库，需提前安装。

import time from selenium import webdriver from bs4 import BeautifulSoup import json

设置浏览器驱动路径

driver_path = 'C:/chromedriver_win32/chromedriver.exe'

创建浏览器对象

browser = webdriver.Chrome(executable_path=driver_path)

访问电影的评论页面

url = 'https://movie.douban.com/subject/25868125/comments?status=P' browser.get(url)

点击“全部”按钮，展开全部评论

all_button = browser.find_element_by_xpath('//*[@id="comments-section"]/div[1]/div[2]/a') all_button.click() time.sleep(1)

获取全部评论页数

soup = BeautifulSoup(browser.page_source, 'html.parser') page_info = soup.select('.comment-page > a')[-2].get_text() total_page = int(page_info)

循环遍历每一页的评论

comments = [] for i in range(0, total_page): # 获取当前页的评论 url = 'https://movie.douban.com/subject/25868125/comments?start={}&limit=20&status=P&sort=new_score'.format(i*20) browser.get(url) soup = BeautifulSoup(browser.page_source, 'html.parser') comment_list = soup.select('#comments > .comment-item')

# 解析评论信息
for comment_item in comment_list:
    comment = {}
    name = comment_item.select('.comment-info > a')[0].get_text().strip()
    time = comment_item.select('.comment-info > span')[0].get_text().strip()
    content = comment_item.select('.short')[0].get_text().strip()
    comment['name'] = name
    comment['time'] = time
    comment['content'] = content
    comments.append(comment)

保存评论数据为json文件

with open('comments.json', 'w', encoding='utf-8') as f: json.dump(comments, f, ensure_ascii=False)

关闭浏览器

browser.quit(