python爬虫利用selenium实现自动翻页爬取某鱼数据的思路详解

from selenium import webdriver
import time
import re

class Douyu(object):

  def __init__(self):
    # 开始时的url
    self.start_url = "https://www.douyu.com/directory/all"
    # 实例化一个Chrome对象
    self.driver = webdriver.Chrome()
    # 用来写csv文件的标题
    self.start_csv = True

  def __del__(self):
    self.driver.quit()

  def get_content(self):
    # 先让程序两秒,保证页面所有内容都可以加载出来
    time.sleep(2)
    item = {}
    # 获取进入下一页的标签
    next_page = self.driver.find_element_by_xpath("//span[text()='下一页']/..")
    # 获取用于判断是否是最后一页的属性
    is_next_url = next_page.get_attribute("aria-disabled")
    # 获取存储信息的所有li标签的列表
    li_list = self.driver.find_elements_by_xpath("//ul[@class='layout-Cover-list']//li")
    
    # 提取需要的数据
    for li in li_list:
      
      item["user-id"] = li.find_element_by_xpath(".//div[@class='DyListCover-userName']").text
      item["img"] = li.find_element_by_xpath(".//div[@class='DyListCover-imgWrap']//img").get_attribute("src")
      item['class-name'] = li.find_element_by_xpath(".//span[@class='DyListCover-zone']").text
      item["click-hot"] = li.find_element_by_xpath(".//span[@class='DyListCover-hot']").text
      item["click-hot"] = re.sub(r'\n','',item['click-hot'])
      
      # 保存数据
      self.save_csv(item)
    
    # 返回是否有下一页和下一页的点击事件的标签,
    return next_page,is_next_url

  def save_csv(self,item):
    # 将提取存放到csv文件中的内容连接为csv格式文件
    str = ','.join([i for i in item.values()])

    with open('./douyu.csv','a',encoding='utf-8') as f:
      if self.start_csv:
        f.write("用户id,image,所属类,点击热度\n")
        self.start_csv = False
      # 将字符串写入csv文件
      f.write(str)
      f.write('\n')
    print("save success")

  def run(self):
    # 启动chrome并定位到相应页面
    self.driver.get(self.start_url)

    while True:
      # 开始提取数据,并获取下一页的元素
      next_page,is_next = self.get_content()
      if is_next!='false':
        break
      # 点击下一页
      next_page.click()

if __name__=='__main__':
  douyu_spider = Douyu()
  douyu_spider.run()