Python爬蟲實(shí)現(xiàn)模擬點(diǎn)擊動(dòng)態(tài)頁(yè)面

#!/usr/bin/python3
# -*- coding:utf-8 -*-
__author__ = 'mayi'
 
"""
動(dòng)態(tài)頁(yè)面的模擬點(diǎn)擊：
  模擬點(diǎn)擊斗魚直播：http://www.douyu.com/directory/all
  爬取每頁(yè)房間名、直播類型、主播名稱、在線人數(shù)等數(shù)據(jù)，然后模擬點(diǎn)擊下一頁(yè)，繼續(xù)爬取
"""
 
from selenium import webdriver
import json
 
# 調(diào)用環(huán)境變量指定的PhantomJS瀏覽器創(chuàng)建瀏覽器對(duì)象，executable_path：指定PhantomJS位置
driver = webdriver.PhantomJS(executable_path = r"D:\Program Files\phantomjs\bin\phantomjs")
from bs4 import BeautifulSoup
 
class DouyuSpider(object):
  """
  爬蟲類
  """
  def __init__(self):
    self.url = "http://www.douyu.com/directory/all/"
    self.driver = webdriver.PhantomJS()
    self.file_name = open("douyu.json", "w", encoding = "utf-8")
 
  def run(self):
    """
    爬蟲開(kāi)始工作
    """
    self.driver.get(self.url)
    # 循環(huán)處理每一頁(yè)，直至最后一頁(yè)
    page = 1
    start_flag = True
    while True:
      # 等待3秒，防止訪問(wèn)過(guò)于頻繁
      self.driver.implicitly_wait(3)
      print("正在處理第" + page + "頁(yè)......")
      page += 1
      # 解析
      soup = BeautifulSoup(self.driver.page_source, "lxml")
      # 在線直播部分
      online_live = soup.find_all('ul', {'id': 'live-list-contentbox'})[0]
      # 房間列表
      live_list = online_live.find_all('li')
      # 處理每一個(gè)房間
      for live in live_list:
        # 房間名、直播類型、主播名稱、在線人數(shù)
        # 房間名
        home_name = live.find_all('h4', {'class': 'ellipsis'})[0].get_text().strip()
        # 直播類型
        live_type = live.find_all('span', {'class': 'tag ellipsis'})[0].get_text().strip()
        # 主播名稱
        anchor_name = live.find_all('span', {'class': 'dy-name ellipsis fl'})[0].get_text().strip()
        # 在線人數(shù)
        online_num = live.find_all('span', {'class' :'dy-num fr'})[0].get_text().strip()
        # print(home_name, live_type, anchor_name, online_num)
        item = {}
        item["房間名"] = home_name
        item["直播類型"] = live_type
        item["主播名稱"] = anchor_name
        item["在線人數(shù)"] = online_num
        if start_flag:
          start_flag = False
          content = "[\n" + json.dumps(item)
        else:
          content = ",\n" + json.dumps(item)
        self.file_name.write(content)
      # page_source.find()未找到內(nèi)容則返回-1
      if self.driver.page_source.find('shark-pager-disable-next') != -1:
        # 已到最后一頁(yè)
        break
      # 模擬點(diǎn)擊下一頁(yè)
      self.driver.find_element_by_class_name('shark-pager-next').click()
 
    # 爬蟲結(jié)束前關(guān)閉文件
    self.file_name.write("\n]")
    self.file_name.close()
if __name__ == '__main__':
  douyu = DouyuSpider()
  douyu.run()