坦白说自动获取有效好友

收到的QQ坦白说通常会提供两条信息，比如‘一个天蝎座的男生’，‘一个认识5年的女生’，那如何找出符合以上条件的好友呢？如果你有几百甚至几千个好友，一个一个找不现实。自动寻找方法当然是通过网络爬虫获取信息再分析过滤实现。
思路：
1. 登录自己QQ空间，获取所有好友QQ号
2. 通过好友QQ号，爬好友空间，获得好友的星座、性别、年龄等个人信息
3. 通过自己QQ号和好友QQ号，获取认识天数，因为认识天数是和两个你和你的好友两个QQ号一起决定的
4. 如果遇到好友空间不让你访问，这个不影响以上信息，因为就算你不能访问好友空间，也可以看见好友的个人信息
5. 如果好友没有填写信息，你当然获取不到相关信息
编程语言：Python
用到的库：
from selenium import webdriver from selenium.webdriver.chrome.options import
Optionsimport time import re import datetime
用到的知识：
1. 通过网址访问云端获取好友信息
2. 用selenium和浏览器配合抓取网页
3. 用正则表达式等处理网页数据，获得有效信息


注意：频繁抓取网页会被腾讯当成违规操作，造成封号两个小时，我的QQ已经被封号两次了，现在还在封号中，通过每获取5个好友信息就退出登录，等两分钟后再登录也许，只是也许可以解决问题。

代码如下（改进登录时间）：
下载代码也可以在链接：https://download.csdn.net/download/gengli2017/10619273
<https://download.csdn.net/download/gengli2017/10619273>
对于初学者，想知道一些代码意思，尽快写一个理解说明。
初学时也找不到资料。
from selenium import webdriver from selenium.webdriver.chrome.options import
Optionsimport time import re import datetime def frankSpeak(account, password) :
myQQ = account driver = loginQQ(account, password) gtk, g_qzonetoken =
getGtk_Token(driver) friendsDict = getFriends(driver, myQQ, gtk, g_qzonetoken)
i =1 friendInfoVector = [] for friendQQ in friendsDict : i += 1 (age, city,
constellation, gender, province) = getFriendInfo(driver, friendQQ, gtk,
g_qzonetoken) knownTime = getKnownTime(driver, myQQ, friendQQ, gtk,
g_qzonetoken) friendInfo = (friendQQ, friendsDict[friendQQ], age, gender,
constellation, knownTime, province, city)
#(age,city,constellation,gender,province) #male:gender=1, famale:gender=0,
else:gender=2 friendInfoVector.append(friendInfo) #to aviod your QQ locked by
Tenser, find 5 friends infomation then reload if(i % 5 == 0) : time.sleep(120)
driver = loginQQ(account, password) gtk, g_qzonetoken = getGtk_Token(driver)
print('一个双鱼座的女生') for i in range(len(friendInfoVector)) : if
(friendInfoVector[i][4] == '双鱼座' and gender == '0') : print(friendInfoVector[i][
1]) print('所有好友信息') print('(QQ号, 备注, 年龄, 性别, 星座, 认识时间, 省份, 城市)') print('女：性别=0
男：性别=1 未注明：性别=2') for i in range(len(friendInfoVector)) :
print(friendInfoVector[i])# print(gtk) # print(g_qzonetoken) #
print(friendsDick) #登录QQ，获取QQ页面 def loginQQ(account, password) : chrome_options
= Options() chrome_options.add_argument("--disable-infobars") driver =
webdriver.Chrome()#这个是chormedriver的地址 driver.get('https://qzone.qq.com/')
driver.switch_to.frame('login_frame') driver.find_element_by_id(
'switcher_plogin').click() driver.find_element_by_id('u').clear()
driver.find_element_by_id('u').send_keys(account) driver.find_element_by_id('p'
).clear() driver.find_element_by_id('p').send_keys(password)
driver.find_element_by_id('login_button').click() time.sleep(2) return driver
#从Cookie获取GTK def getGTKFromCookie(cookie): hashes = 5381 for letter in cookie[
'p_skey']: hashes += (hashes << 5) + ord(letter) return hashes & 0x7fffffff
#获取gtk和g_qzonetoken def getGtk_Token(driver) : cookie = {} #初始化cookie字典 for elem
in driver.get_cookies(): #取cookies cookie[elem['name']] = elem['value'] gtk =
getGTKFromCookie(cookie) html = driver.page_source g_qzonetoken=re.search(
'window\.g_qzonetoken = \(function\(\)\{ try\{return (.*?);\} catch\(e\)',html)
#从网页源码中提取g_qzonetoken g_qzonetoken = str(g_qzonetoken[0]).split('\"')[1] return
gtk, g_qzonetoken#获得好友列表 def getFriends(driver, myQQ, gtk, g_qzonetoken) :
friendUrl =
'https://user.qzone.qq.com/proxy/domain/r.qzone.qq.com/cgi-bin/tfriend/friend_hat_get.cgi?hat_seed=1&uin='
+ str(myQQ) +'&fupdate=1&g_tk='+str(gtk)+'&qzonetoken='+str(g_qzonetoken)+
'&g_tk='+str(gtk) driver.get(friendUrl) friend_list = driver.page_source
friend_list = str(friend_list ) pattern = re.compile(
'\"(.\d*)\":\{\\n"realname":"(.*?)"}',re.S) QQ_name_list = re.findall(pattern,
str(friend_list)) friendDick=dict()#numList => (QQnum:QQname) for friend in
QQ_name_list: friendDick[str(friend[0])]=str(friend[1]) return friendDick
#通过好友QQ号获取好友信息 def getFriendInfo(driver, friendQQ, gtk, g_qzonetoken) :
friendInfoUrl ='https://mobile.qzone.qq.com/profile_get?qzonetoken='
+str(g_qzonetoken) +'&g_tk='+str(gtk)+'&format=json&hostuin=' + str(friendQQ)
driver.get(friendInfoUrl) friendInfo = driver.page_source friendInfo =
str(friendInfo)#savefile(str(friendQQ)+'txt', friendInfo) pattern = re.compile(
r'"age":(\d*).*"city":"(\w*)".*"constellation":"(\w*).*"gender":(-?\d*).*"province":"(\w*)"'
) usefulInfo = pattern.findall(friendInfo)if (any(usefulInfo)) : return
usefulInfo[0] else : return ('-1','NULL','NULL','2','NULL')
#usefulInfo=[(age,city,constellation,gender,province)]
#usefulInfo[0]=(age,city,constellation,gender,province) #male:gender=1,
famale:gender=0, else:gender=2 #通过自己QQ和好友QQ获取认识的时间 def getKnownTime(driver,
myQQ, friendQQ, gtk, g_qzonetoken) : knownDaysUrl =
'https://user.qzone.qq.com/proxy/domain/r.qzone.qq.com/cgi-bin/friendship/cgi_friendship?activeuin='
+ str(myQQ) +'&passiveuin=' + str(friendQQ) +'&situation=1&isCalendar=1&g_tk='
+str(gtk) +'&qzonetoken='+str(g_qzonetoken)+'&g_tk='+str(gtk)
driver.get(knownDaysUrl) knownDaysInfo = driver.page_source knownDaysInfo =
str(knownDaysInfo) beginStamp = re.search(r'"addFriendTime":(\d+)',
knownDaysInfo) beginStamp = str(beginStamp.group(1)) beginTime =
datetime.date.fromtimestamp(int(beginStamp)) beginY = beginTime.year beginM =
beginTime.month beginD = beginTime.day lastStamp = re.search(
r'"systemTime":(\d+)', knownDaysInfo) lastStamp = str(lastStamp.group(1))
lastTime = datetime.date.fromtimestamp(int(lastStamp)) lastY = lastTime.year
lastM = lastTime.month lastD = lastTime.dayif(lastY > beginY) : return
(str(lastY - beginY +1) + '年') elif(lastM > lastM) : return (str(lastM - beginM
+1) + '月') else : return (str(lastD - beginD + 1) + '日' #运行程序，把QQNumber换成要登录QQ，
password换成密码 frankSpeak('QQNumber', 'password')

• « 上一篇：Kotlin中常量和静态方法
• » 下一篇：Python 爬取蔡徐坤的 10 万转发数据，竟是假流量？