标签:XML from 数据 bre dump list pager json __init__
动态页面的模拟点击:
以斗鱼直播为例:http://www.douyu.com/directory/all
爬取每页的房间名、直播类型、主播名称、在线人数等数据,然后模拟点击下一页,继续爬取
#!/usr/bin/python3 # -*- conding:utf-8 -*- __author__ = ‘mayi‘ """ 动态页面的模拟点击: 模拟点击斗鱼直播:http://www.douyu.com/directory/all 爬取每页房间名、直播类型、主播名称、在线人数等数据,然后模拟点击下一页,继续爬取 """ from selenium import webdriver import json # 调用环境变量指定的PhantomJS浏览器创建浏览器对象,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): """ 爬虫类 """ 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): """ 爬虫开始工作 """ self.driver.get(self.url) # 循环处理每一页,直至最后一页 page = 1 start_flag = True while True: # 等待3秒,防止访问过于频繁 self.driver.implicitly_wait(3) print("正在处理第" + page + "页......") page += 1 # 解析 soup = BeautifulSoup(self.driver.page_source, "lxml") # 在线直播部分 online_live = soup.find_all(‘ul‘, {‘id‘: ‘live-list-contentbox‘})[0] # 房间列表 live_list = online_live.find_all(‘li‘) # 处理每一个房间 for live in live_list: # 房间名、直播类型、主播名称、在线人数 # 房间名 home_name = live.find_all(‘h3‘, {‘class‘: ‘ellipsis‘})[0].get_text().strip() # 直播类型 live_type = live.find_all(‘span‘, {‘class‘: ‘tag ellipsis‘})[0].get_text().strip() # 主播名称 anchor_name = live.find_all(‘span‘, {‘class‘: ‘dy-name ellipsis fl‘})[0].get_text().strip() # 在线人数 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["房间名"] = home_name item["直播类型"] = live_type item["主播名称"] = anchor_name item["在线人数"] = 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()未找到内容则返回-1 if self.driver.page_source.find(‘shark-pager-disable-next‘) != -1: # 已到最后一页 break # 模拟点击下一页 self.driver.find_element_by_class_name(‘shark-pager-next‘).click() # 爬虫结束前关闭文件 self.file_name.write("\n]") self.file_name.close() if __name__ == ‘__main__‘: douyu = DouyuSpider() douyu.run()
标签:XML from 数据 bre dump list pager json __init__
原文地址:http://www.cnblogs.com/mayi0312/p/7236472.html