QQ空间动态内容,好友信息,点赞爬虫脚本
一、安装基础的软件包:
1、准备好火狐浏览器,并下载geckodriver,将geckodriver加入到环境变量:
下载geckodriver的地址:https://pan.baidu.com/s/1NDo04Zj8NxmlRe90_CDRow
2、下载selenium:
python pip install selenium
二、脚本说明:
from selenium import webdriver from time import sleep username='xxxxxxxxx' password='xxxxxxxxxx' qzone_url ='https://qzone.qq.com/' '''qq空间自动登录登录''' def login_qzone(url,username,password,browser_type='Firefox'): if browser_type == 'Firefox': driver = webdriver.Firefox() elif browser_type == 'Firefox_headless': #用火狐的无边界浏览器 fireFoxOptions = webdriver.FirefoxOptions() fireFoxOptions.set_headless() driver = webdriver.Firefox(firefox_options=fireFoxOptions) driver.get(url) if 'qzone' in url: pass else: driver.find_element_by_class_name('qzone').click() sleep(5) driver.switch_to.window(driver.window_handles[1]) #print(driver.window_handles) driver.switch_to.frame('login_frame') #这个很坑,里面还包括子页面 pwd_login_button = driver.find_element_by_id('switcher_plogin') pwd_login_button.click() #点击通过密码登录 input_user = driver.find_element_by_id('u') input_user.send_keys(username) input_pwd = driver.find_element_by_name('p') input_pwd.send_keys(password) login_button = driver.find_element_by_id('login_button') login_button.click() driver.switch_to.default_content() sleep(5) return driver if __name__ == '__main__': login_qzone(qzone_url,username,password)
import qzone_login as login from selenium.common.exceptions import StaleElementReferenceException,NoSuchElementException from selenium import webdriver import time import json qq_number_dict ={} def save_qq_number(tag_list): for line in tag_list: qq_number = line.get_attribute('href').split('/')[-1] qq_number_dict.update({qq_number:line.text}) '''得到qq空间里的所有好友,保存到文件里''' def save_in_file(): qq_number_dict.update({'save_time':time.time()}) browser = login.login_qzone(login.qzone_url,login.username,login.password,browser_type = 'Firefox_headless') #browser = login.login_qzone(login.qzone_url,login.username,login.password) browser.get('https://user.qzone.qq.com/949885111/myhome/friends/index') browser.switch_to.default_content() frame = browser.find_element_by_tag_name('iframe') browser.switch_to.frame(frame) for i in range(60): target = browser.find_element_by_class_name('qz-button.btn-pager-next') browser.execute_script("arguments[0].scrollIntoView();", target) #屏幕转到qz-button.btn-pager-next类的标签 save_qq_number(browser.find_elements_by_class_name('textoverflow')) target.click() time.sleep(1) with open('qq_friends.json', 'w', encoding='utf-8') as f: f.write(json.dumps(qq_number_dict)) if __name__ == "__main__": with open('qq_friends.json', 'r', encoding='utf-8') as f: for line in f: save_time =json.loads(line)['save_time'] if time.time() - save_time > 86400: save_in_file() else: with open('qq_friends.json', 'r', encoding='utf-8') as f: for line in f: print(len(json.loads(line)))
import qzone_login as login from selenium import webdriver from selenium.webdriver.common.keys import Keys from selenium.common.exceptions import ElementNotInteractableException,NoSuchFrameException,NoSuchElementException import time friend_qzone_url = 'https://user.qzone.qq.com/2453294057' '''判断是否登录成功''' def judge_login(): pass '''进入frame''' def switch_frame(broeser,frame): while True: try: broeser.switch_to.frame(frame) #一共两层 except NoSuchFrameException: pass else: break time.sleep(2) # 睡几秒,不然抓不到frame '''抓异常并初步处理数据''' def get_erroy(browser,tag,return_sign = False): try: execute_output = browser.find_element_by_class_name(tag) except NoSuchElementException : print(111111111111111111111122222) execute_output = '' else: if return_sign == False: execute_output = execute_output.text.replace(' ','').replace('\n','') return execute_output ''''分析生成了的所有页面''' def analyse_html(broswser): data_dict = {} friend_speaks = broswser.find_elements_by_class_name('f-single.f-s-s') for speak in friend_speaks: qzone_talk = get_erroy(speak,'f-info') deta_time = get_erroy(speak,'ui-mr8.state') phone_model_tag = get_erroy(speak,'item',return_sign= True) if phone_model_tag != '': phone_model = phone_model_tag.find_element_by_tag_name('a').text else: phone_model = '' print(qzone_talk,phone_model,deta_time) data_dict.update({deta_time: [phone_model, qzone_talk]}) return data_dict '''判断是否有亲密度弹窗''' def judge_pop_up(browser): if len(browser.find_elements_by_class_name('btn-fs-sure')) != 0: browser.find_elements_by_class_name('btn-fs-sure')[0].click() '''得到好友动态页面的所有数据''' def get_all_data(friend_url): #driver = login.login_qzone(login.qzone_url, login.username, login.password) driver = login.login_qzone(login.qzone_url,login.username,login.password,browser_type = 'Firefox_headless') driver.get(friend_url) driver.switch_to.default_content() time.sleep(4) #睡几秒,不然抓不到亲密度弹窗 judge_pop_up(driver) driver.find_element_by_id("aOwnerFeeds").click() driver.switch_to.default_content() switch_frame(driver,'app_canvas_frame') #进入第一层frame switch_frame(driver,'frameFeedList') #进入好友动态frame while True: try: target = driver.find_element_by_class_name('b-inline.data_btn_more') target.click() except ElementNotInteractableException: try: target = driver.find_element_by_class_name('b-inline.data_is_loading') target.click() except ElementNotInteractableException: pass time.sleep(3) finally: driver.execute_script("arguments[0].scrollIntoView(false);", target) finally_sign = driver.find_elements_by_class_name('b-inline.data_no_more.none') # 找不到这个标签时就退出 print(finally_sign) if len(finally_sign) == 0: break time.sleep(2) return driver '''点赞''' def give_like(browser): for like in browser.find_elements_by_class_name('item.qz_like_btn_v3'): browser.execute_script("arguments[0].scrollIntoView(false);", like) like.click() time.sleep(60) if __name__ == '__main__': driver = get_all_data(friend_qzone_url) data = analyse_html(driver) print(data)
上面三个脚本包括:
1、第一个是登录的脚本,可以选择使用火狐的无边界模式。
2、第二个使用第一个登录后,将空间里的QQ好友信息拿到(qq号:好友备注)保存到json文件里面。
3、第三个使用第一个登录后,有两个功能:
1.得到该好友的所有动态的内容(只包括说说内容,发表的时间,和使用的手机号)
2.可以给好友点赞。注:点太快了会被冻结
注:以上只是学习selenium所写的小脚本,可别用于非法用途。