利用cookielib和urllib2模块模拟登陆163的例子有很多,近期看了《python模拟登陆163邮箱并获取通讯录》一文,受到启发,试着对收件箱、发件箱等进行了分析,并列出了所有邮件列表及状态,包括发件人、收件人、主题、发信时间、已读未读等状态。
1、参考代码:http://hi.baidu.com/fc_lamp/blog/item/2466d1096fcc532de8248839.html%EF%BB%BF
#-*- coding:UTF-8 -*- import urllib,urllib2,cookielib import xml.etree.ElementTree as etree #xml解析类 class Login163: #伪装browser header = {'User-Agent':'Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.1.6) Gecko/20091201 Firefox/3.5.6'} username = '' passwd = '' cookie = None #cookie对象 cookiefile = './cookies.dat' #cookie临时存放地 user = '' def __init__(self,username,passwd): self.username = username self.passwd = passwd #cookie设置 self.cookie = cookielib.LWPCookieJar() #自定义cookie存放 opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(self.cookie)) urllib2.install_opener(opener) #登陆 def login(self): #请求参数设置 postdata = { 'username':self.username, 'password':self.passwd, 'type':1 } postdata = urllib.urlencode(postdata) #发起请求 req = urllib2.Request( url='http://reg.163.com/logins.jsp?type=1&product=mail163&url=http://entry.mail.163.com/coremail/fcg/ntesdoor2?lightweight%3D1%26verifycookie%3D1%26language%3D-1%26style%3D1', data= postdata,#请求数据 headers = self.header #请求头 ) result = urllib2.urlopen(req).read() result = str(result) self.user = self.username.split('@')[0] self.cookie.save(self.cookiefile)#保存cookie if '登录成功,正在跳转...' in result: #print("%s 你已成功登陆163邮箱。---------\n" %(user)) flag = True else: flag = '%s 登陆163邮箱失败。'%(self.user) return flag #获取通讯录 def address_list(self): #获取认证sid auth = urllib2.Request( url='http://entry.mail.163.com/coremail/fcg/ntesdoor2?username='+self.user+'&lightweight=1&verifycookie=1&language=-1&style=1', headers = self.header ) auth = urllib2.urlopen(auth).read() for i,sid in enumerate(self.cookie):#enumerate()用于同时返数字索引与数值,实际上是一个元组:((0,test[0]),(1,test[1]).......)这有点像php里的foreach 语句的作用 sid = str(sid) if 'sid' in sid: sid = sid.split()[1].split('=')[1] break self.cookie.save(self.cookiefile) #请求地址 url = 'http://twebmail.mail.163.com/js4/s?sid='+sid+'&func=global:sequential&showAd=false&userType=browser&uid='+self.username #参数设定(var 变量是必需要的,不然就只能看到:<code>S_OK</code><messages/>这类信息) #这里参数也是在firebug下查看的。 postdata = { 'func':'global:sequential', 'showAd':'false', 'sid':sid, 'uid':self.username, 'userType':'browser', 'var':'<?xml version="1.0"?><object><array name="items"><object><string name="func">pab:searchContacts</string><object name="var"><array name="order"><object><string name="field">FN</string><boolean name="desc">false</boolean><boolean name="ignoreCase">true</boolean></object></array></object></object><object><string name="func">pab:getAllGroups</string></object></array></object>' } postdata = urllib.urlencode(postdata) #组装请求 req = urllib2.Request( url = url, data = postdata, headers = self.header ) res = urllib2.urlopen(req).read() #解析XML,转换成json #说明:由于这样请求后163给出的是xml格式的数据, #为了返回的数据能方便使用最好是转为JSON json = [] tree = etree.fromstring(res) obj = None for child in tree: if child.tag == 'array': obj = child break #这里多参考一下,etree元素的方法属性等,包括attrib,text,tag,getchildren()等 obj = obj[0].getchildren().pop() for child in obj: for x in child: attr = x.attrib if attr['name']== 'EMAIL;PREF': value = {'email':x.text} json.append(value) return json #Demo print("Requesting......\n\n") login = Login163('xxxx@163.com','xxxxx') flag = login.login() if type(flag) is bool: print("Successful landing,Resolved contacts......\n\n") res = login.address_list() for x in res: print(x['email']) else: print(flag)
2、分析收件箱、发件箱等网址
在参考代码中,获取通讯录的url为
url = 'http://twebmail.mail.163.com/js4/s?sid='+sid+'&func=global:sequential&showAd=false&userType=browser&uid='+self.username,通过对邮箱地址的分析,发现收件箱、发件箱等的url为url = 'http://twebmail.mail.163.com/js4/s?sid='+sid+'&func=mbox:listMessages&showAd=false&userType=browser&uid='+self.username,其中func=
mbox:listMessages。其对收件箱、发件箱的具体区分在下面的postdata中,具体为:
(1)收件箱
postdata = { 'func':'global:sequential', 'showAd':'false', 'sid':'qACVwiwOfuumHPdcYqOOUTAjEXNbBeAr', 'uid':self.username, 'userType':'browser', 'var':'<!--?xml version="1.0"?--><object><int name="fid">1</int><string name="order">date</string><boolean name="desc">true</boolean><boolean name="topFirst">false</boolean><int name="start">0</int><int name="limit">20</int></object>' }
(2)发件箱
postdata = { 'func':'global:sequential', 'showAd':'false', 'sid':'qACVwiwOfuumHPdcYqOOUTAjEXNbBeAr', 'uid':self.username, 'userType':'browser', 'var':'<!--?xml version="1.0"?--><object><int name="fid">3</int><string name="order">date</string><boolean name="desc">true</boolean><boolean name="topFirst">false</boolean><int name="start">0</int><int name="limit">20</int></object>' }
可以看出,两段代码的不同之处就是fid的取值不同,其中收件箱为1,发件箱为3,草稿箱为2。
3、xml解析
利用ElementTree 类来进行xml到字典的转换。在获取通讯录的实例中,主要使用了这一方法。本例子(具体代码见后文)在收取邮件列表时,并没有用这一方法,仍然使用的是字符串的处理方法。但这里还是列一下ElementTree 类对xml的处理。如(参考地址:http://hi.baidu.com/fc_lamp/blog/item/8ed2d53ada4586f714cecb3d.html):
-<result> <code>S_OK</code> -<array name="var"> -<object> <string name="code">S_OK</string> -<array name="var"> +<object></object> +<object></object> +<object></object> +<object></object> +<object></object> +<object></object> +<object></object> +<object></object> +<object></object> +<object></object> +<object></object> +<object></object> +<object></object> +<object></object> +<object></object> +<object></object> </array> </object> +<object></object> </array> </result>
解决方法:
#-*- coding:UTF-8 -*- import xml.etree.ElementTree as etree #xml解析类 def xml2json(xml): json = [] tree = etree.fromstring(xml) #如果是文件可用parse(source) obj = None for child in tree: if child.tag == 'array': obj = child break #这里多参考一下,etree元素的方法属性等,包括attrib,text,tag,getchildren()等 obj = obj[0].getchildren().pop() for child in obj: for x in child: attr = x.attrib if attr['name']== 'EMAIL;PREF': value = {'email':x.text} json.append(value) return json
4、收件箱邮件列表
本例子只列出了收件箱邮件列表,如果需要,可根据以上介绍调整fid值,列出发件箱、草稿箱等的邮件列表。程序在windosxp、py2.6环境下调查通过,运行后,会在当前目录下生成三个文件:inboxlistfile.txt记录收件箱邮件列表,addfile.txt记录通讯录,cookies.dat记录cookies。具体代码如下:
#-*- coding:UTF-8 -*- #@小五义 http://www.cnblogs.com/xiaowuyi #163邮件列表 import urllib,urllib2,cookielib import xml.etree.ElementTree as etree #xml解析类 class Login163: #伪装browser header = {'User-Agent':'Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.1.6) Gecko/20091201 Firefox/3.5.6'} username = '' passwd = '' cookie = None #cookie对象 cookiefile = './cookies.dat' #cookie临时存放地 user = '' def __init__(self,username,passwd): self.username = username self.passwd = passwd #cookie设置 self.cookie = cookielib.LWPCookieJar() #自定义cookie存放 opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(self.cookie)) urllib2.install_opener(opener) #登陆 def login(self): #请求参数设置 postdata = { 'username':self.username, 'password':self.passwd, 'type':1 } postdata = urllib.urlencode(postdata) #发起请求 req = urllib2.Request( url='http://reg.163.com/logins.jsp?type=1&product=mail163&url=http://entry.mail.163.com/coremail/fcg/ntesdoor2?lightweight%3D1%26verifycookie%3D1%26language%3D-1%26style%3D1', data= postdata,#请求数据 headers = self.header #请求头 ) result = urllib2.urlopen(req).read() result = str(result) #print result self.user = self.username.split('@')[0] self.cookie.save(self.cookiefile)#保存cookie if '登录成功,正在跳转...' in result: #print("%s 你已成功登陆163邮箱。---------n" %(user)) flag = True else: flag = '%s 登陆163邮箱失败。'%(self.user) return flag #获取通讯录 def address_list(self): #获取认证sid auth = urllib2.Request( url='http://entry.mail.163.com/coremail/fcg/ntesdoor2?username='+self.user+'&lightweight=1&verifycookie=1&language=-1&style=1', headers = self.header ) auth = urllib2.urlopen(auth).read() #authstr=str(auth) #print authstr for i,sid in enumerate(self.cookie): sid = str(sid) #print 'sid:%s' %sid if 'sid' in sid: sid = sid.split()[1].split('=')[1] break self.cookie.save(self.cookiefile) #请求地址 url = 'http://twebmail.mail.163.com/js4/s?sid='+sid+'&func=global:sequential&showAd=false&userType=browser&uid='+self.username #参数设定(var 变量是必需要的,不然就只能看到:<code>S_OK</code><messages>这类信息) #这里参数也是在firebug下查看的。 postdata = { 'func':'global:sequential', 'showAd':'false', 'sid':'qACVwiwOfuumHPdcYqOOUTAjEXNbBeAr', 'uid':self.username, 'userType':'browser', 'var':'<!--?xml version="1.0"?--><object><array name="items"><object><string name="func">pab:searchContacts</string><object name="var"><array name="order"><object><string name="field">FN</string><boolean name="desc">false</boolean><boolean name="ignoreCase">true</boolean></object></array></object></object><object><string name="func">pab:getAllGroups</string></object></array></object>' } postdata = urllib.urlencode(postdata) #组装请求 req = urllib2.Request( url = url, data = postdata, headers = self.header ) res = urllib2.urlopen(req).read() #print str(res) #解析XML,转换成json #说明:由于这样请求后163给出的是xml格式的数据, #为了返回的数据能方便使用最好是转为JSON json = [] tree = etree.fromstring(res) obj = None for child in tree: if child.tag == 'array': obj = child break #这里多参考一下,etree元素的方法属性等,包括attrib,text,tag,getchildren()等 obj = obj[0].getchildren().pop() for child in obj: for x in child: attr = x.attrib if attr['name']== 'EMAIL;PREF': value = {'email':x.text} json.append(value) return json def minbox(self):#收件箱,fid为1,发件箱为3,草稿箱为2 #获取认证sid auth = urllib2.Request( url='http://entry.mail.163.com/coremail/fcg/ntesdoor2?username='+self.user+'&lightweight=1&verifycookie=1&language=-1&style=1', headers = self.header ) auth = urllib2.urlopen(auth).read() #authstr=str(auth) #print authstr for i,sid in enumerate(self.cookie): sid = str(sid) #print 'sid:%s' %sid if 'sid' in sid: sid = sid.split()[1].split('=')[1] break self.cookie.save(self.cookiefile) #请求地址 url = 'http://twebmail.mail.163.com/js4/s?sid='+sid+'&func=mbox:listMessages&showAd=false&userType=browser&uid='+self.username #参数设定(var 变量是必需要的,不然就只能看到:<code>S_OK</code><messages>这类信息) #这里参数也是在firebug下查看的。 postdata = { 'func':'global:sequential', 'showAd':'false', 'sid':'qACVwiwOfuumHPdcYqOOUTAjEXNbBeAr', 'uid':self.username, 'userType':'browser', 'var':'<!--?xml version="1.0"?--><object><int name="fid">1</int><string name="order">date</string><boolean name="desc">true</boolean><boolean name="topFirst">false</boolean><int name="start">0</int><int name="limit">20</int></object>' } postdata = urllib.urlencode(postdata) #组装请求 req = urllib2.Request( url = url, data = postdata, headers = self.header ) res = urllib2.urlopen(req).read() liststr=str(res).split('<object>')#用object进行分割 inboxlistcount=len(liststr)-1#记录邮件封数 inboxlistfile=open('inboxlistfile.txt','a') t=0 #记录当前第几封信 for i in liststr: if 'xml' in i and ' version=' in i: inboxlistfile.write('inbox 共'+str(inboxlistcount)+'信') inboxlistfile.write('\n') if 'name="id"' in i: t=t+1 inboxlistfile.write('第'+str(t)+'封:') inboxlistfile.write('\n') #写入from beginnum=i.find('name="from"') endnum=i.find('</string>',beginnum) inboxlistfile.write('From:'+i[beginnum+12:endnum]) inboxlistfile.write('\n') #写入to beginnum=i.find('name="to"') endnum=i.find('</string>',beginnum) inboxlistfile.write('TO:'+i[beginnum+10:endnum]) inboxlistfile.write('\n') #写入subject beginnum=i.find('name="subject"') endnum=i.find('</string>',beginnum) inboxlistfile.write('Subject:'+i[beginnum+15:endnum]) inboxlistfile.write('\n') #写入date: beginnum=i.find('name="sentDate"') endnum=i.find('</date>',beginnum) inboxlistfile.write('Date:'+i[beginnum+16:endnum]) inboxlistfile.write('\n') if 'name="read">true' in i: inboxlistfile.write('邮件状态:已读') inboxlistfile.write('\n') else: inboxlistfile.write('邮件状态:未读') inboxlistfile.write('\n') #写用邮件尺寸 beginnum=i.find('name="size"') endnum=i.find('</int>',beginnum) inboxlistfile.write('邮件尺寸:'+i[beginnum+12:endnum]) inboxlistfile.write('\n') #写入邮件编号,用于下载邮件 beginnum=i.find('name="id"') endnum=i.find('</string>',beginnum) inboxlistfile.write('邮件编号:'+i[beginnum+10:endnum]) inboxlistfile.write('\n\n') inboxlistfile.close() #Demo print("Requesting......nn") login = Login163('AAAAA@163.com','AAAAA') flag = login.login() if type(flag) is bool: login.minbox() #login.letterdown() print("Successful landing,Resolved contacts......nn") res = login.address_list() for x in res: print(x['email']) else: print(flag)