#此程序用来爬取网易云音乐中《将故事写成我们》的评论，主要包含用户名和评论
#其中csv文件保存了用户名和评论，txt文件只保存了评论
from selenium import webdriver
import time
import csv
import bs4
from selenium.webdriver.common.by import By
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.support.wait import WebDriverWait
url='https://music.163.com/#/song?id=1391477429'
#翻页的函数
def next_page():ckl = driver.find_element_by_link_text('下一页')driver.execute_script("window.scrollTo(0,document.body.scrollHeight)")ckl.click()wait = WebDriverWait(driver, 10)wait.until(EC.presence_of_element_located((By.CLASS_NAME,'itm')))content = driver.page_sourcecontent = content.replace('<br />', 'n')  # 输出时按照网页上的格式换行保存评论内容html = bs4.BeautifulSoup(content, 'lxml')return html
if __name__ == '__main__':driver = webdriver.Chrome()driver.get(url)time.sleep(5)list_info = []time.sleep(1)#首先找到g_iframe这个框架，不然直接找找不到frame = driver.find_element_by_id('g_iframe')driver.switch_to.frame(frame)time.sleep(2)#翻页是因为第一次直接显示最火的15条评论，翻页再翻回去就没有那15条评论了# 先翻一页sec=driver.find_element_by_link_text('下一页')driver.execute_script("window.scrollTo(0,document.body.scrollHeight)")sec.click()time.sleep(1)#再回到第一页fir=driver.find_element_by_link_text('上一页')fir.click()time.sleep(1)# 先爬取首页的评论datas = driver.find_elements_by_css_selector('div.cntwrap > div:nth-child(1) > div')for i in datas:data=i.textname=data[:data.index('：')]comment=data[data.index('：')+1:]list_info.append([name,comment])# print(list_info)# 爬取后续的页面评论for i in range(200):time.sleep(0.5)ht=next_page()ncom=ht.find_all('div',class_='cnt f-brk')for k in ncom:data1=k.textname1=data1[:data1.index('：')]comment1=data1[data1.index('：')+1:]list_info.append([name1,comment1])print(list_info)#将其保存成csv文件with open('将故事写成我们','a') as f:writer=csv.writer(f)writer.writerows(list_info)#将其中的评论comment部分保存成txt文件with open('将故事写成我们的评论.txt','a') as f:for i in list_info:f.write(i[1])list_info=[]print('Done')