Python 之selenium+phantomJS斗鱼抓取案例

from selenium import webdriver
from bs4 import BeautifulSoup
import time

if __name__ == '__main__':
    driver = webdriver.PhantomJS()
    driver.get("https://www.douyu.com/directory/all")
    count = 0  # 总页数
    total = 0  # 房间数
    time.sleep(5)
    while True:

        # 判断是否到达尾页，如果是true，则推出循环
        attr = driver.find_element_by_css_selector(".dy-Pagination-next").get_attribute("aria-disabled")
        if "true" in attr:
            break
        try:
            page_count = 0 # 记录每页多少条数据
            # 解析html
            soup = BeautifulSoup(driver.page_source, "lxml")
            names = soup.find_all("h2", attrs={"class": "DyListCover-user"})
            looks = soup.find_all("span", attrs={"class": "DyListCover-hot"})
            for name, look in zip(names, looks):
                print("房间：" + name.get_text() + "\t人数：" + look.get_text())
                total += 1
                page_count += 1
            count += 1
            print("==================第%s页，每页%s条数据==================" % (str(count), str(page_count)))
            # 保存每一页的图片
            driver.save_screenshot("./douyu/douyu_%s.png" % str(count))
            # 点击下一页
            driver.find_element_by_css_selector(".dy-Pagination-next").click()
            # 等待数据加载完毕
            time.sleep(3)
        except RuntimeError as e:
            print(e)
    print("总页数：" + str(count))
    print("总房间数：" + str(total))

结果如图：