1. # https://blog.csdn.net/qq_40821402/article/details/88178359
    3. from lxml import html
    4. import time
    5. from selenium import webdriver
    6. import re
    7. import csv
    8. import pymysql
    9. class Demo:
    11.     comment_list = []  # 定义评论列表文件
    12.     writer_list = []  # 定义评论者列表
    13.     # 写入数据库
    14.     conn = pymysql.connect(host='localhost' ,user = 'root',passwd =  'root',db= 'wyy' ,charset = 'utf8mb4')
    15.     curs = conn.cursor()
    18.     def __init__(self,browser,url):
    19.         self.browser = browser
    20.         time.sleep(2)
    21.         self.browser.get(url)
    22.         self.browser.implicitly_wait(65)
    23.         try:
    24.             self.browser.switch_to.frame('contentFrame')  # 进入frame
    25.         except Exception as e:
    26.             print(e)
    28.     def all_pages(self):
    29.         etree = html.etree
    30.         htmls = self.browser.page_source
    31.         codes = etree.HTML(htmls)
    32.         texts = codes.xpath("//a[@href='#']")
    33.        # for te in texts:   #记录一下总的文本数量13个
    34.             #print(te.text)
    35.         s = 4
    36.         while (s<=13):
    37.             pa = texts[len(texts) - s].text
    38.             if pa == None:
    39.                 s = s+1
    40.             else:
    41.                 last_page = int(pa)
    42.                 return last_page
    44.     def get_comment(self, pages):
    45.         for x in range(pages):  #64是需要抓取的页面总数***
    46.             button = self.browser.find_element_by_link_text('下一页')  # 通过找到‘下一页’的按钮进行翻页
    47.             time.sleep(2)
    48.             # print(button.text)
    49.             aa = self.browser.find_elements_by_xpath('//div/div[2]/div[1]/div[1]')  # 定位评论文字
    50.             for i in aa:
    51.                 # print(i.text)
    52.                 a = re.search('：', i.text)  # 通过正则表达式进行“：”定位，得出左边是评论者，这个要去掉
    53.                 # print(a)
    54.                 if a:
    55.                     n = i.text.index('：') + 1
    56.                     writ = i.text[:n].replace(':', '')  # 只保留从“：”左边的文字，就是纯正的作者名
    57.                     comm = i.text[n:]  # 只保留从“：”右边的文字，就是纯正的评论文字
    58.                     self.curs.execute("insert into data (writer,comment) values (%s,%s) " ,(writ, comm))
    59.                     self.conn.commit()
    60.                 else:
    61.                     #comment_list.append('')
    62.                     print('')
    63.             self.conn.commit()
    64.             print('抓取了>>>>' + str(x+1) + '<<<<页')
    65.             try :
    66.                 button.click()  # 自动进行“下一页”的按钮
    67.             except Exception as e:
    68.                 print(e)
    70. if __name__ == '__main__':
    71.     id = open('F:/music/网易云音乐歌曲信息.csv', 'r', encoding='utf-8-sig')
    72.     id_reader = csv.DictReader(id)
    73.     columns = [row for row in id_reader]
    74.     #columns = [439646311,1348918118,541076059,1303464240]  # 测试切换id
    75.     chrome_options = webdriver.ChromeOptions()
    76.     prefs = {"profile.managed_default_content_settings.images": 2}
    77.     chrome_options.add_experimental_option("prefs", prefs)
    78.     browser = webdriver.Chrome(options= chrome_options)
    79.     for id_new in columns:
    80.         new_id = id_new['id']
    81.         url = 'https://music.163.com/#/song?id=' + str(new_id) # 测试的时候换成id_new
    82.         dd = Demo(browser=browser, url = url)
    83.         pages = dd.all_pages()
    84.         dd.get_comment(pages)
    85.     print('********************************进入下一个************************************')
    86.     time.sleep(3)
    87.     dd.browser.close()

    效果图
    image.png