Day 03(爬取豌豆荚app内容)

import  time
import requests
from selenium import webdriver
from selenium.webdriver.common.keys import Keys
'''

游戏名称
游戏大小
下载人数
游戏简介
下载地址

'''
def get_app(driver):

        app_list = driver.find_elements_by_class_name('card')
        for app in app_list:
            app_name = app.find_element_by_class_name('app-title-h2').text
            app_count = app.find_element_by_class_name('install-count').text
            app_size = app.find_element_by_css_selector('.meta span').get_attribute('title')
            app_com = app.find_element_by_class_name('comment').text
            app_url = app.find_element_by_css_selector('.icon-wrap a').get_attribute('href')

            app_content = f'''
                游戏名称:{app_name}
                游戏大小:{app_size}
                下载人数:{app_count}
                游戏简介:{app_com}
                下载地址:{app_url}
            '''
            print(app_content)

            with open('wdj.text','a',encoding='utf-8')as f:
                f.write(app_content)

        print('导入成功...')

if __name__ == '__main__':
    try:
        driver = webdriver.Chrome('E:\chromedriver_win32\chromedriver.exe')
        driver.implicitly_wait(5)
        driver.get('https://www.wandoujia.com/category/6001')

        get_app(driver)
    finally:
        driver.close()

猜你喜欢

转载自www.cnblogs.com/zzx12181/p/11127849.html