def loginTom(username, password):
 url1 = '''
 http://login.mail.tom.com/cgi/login
 '''

 values = {
  'type' : '0',
  'user' : '%s' % username,
  'in_username' : '%s@tom.com' % username,
  'pass' : '%s' % password,
  'style' : '21',
  'verifycookie' : 'y'
 }

 data = urllib.urlencode(values)
 req = urllib2.Request(url1, data)
 response = opener.open(req)
 data2 =response.read()

 sid = re.search(r'(?<=(sid=)).*?(?=&)', data2).group()

 url3 = '''
 http://bjapp6.mail.tom.com/cgi/ldapapp?funcid=mails&sid=%s&fid=1
 ''' % sid
 response = opener.open(url3)
 data3 = response.read()

 b = re.search(r'(?<=nTotalMailCount).*?(?=;)', data3).group()
 c = re.search(r'/d.+', b).group()
 num_per_page = 20
 num_times = string.atoi(c) / 20
 print( num_times )

 index = 0
 for match in re.finditer(r'(?<="Mbox_Td_Subject"/>).*?(?=/</)', data3):
  index += 1
  part1 = "第%d封郵件" % index
  part2 = " %s" % match.group()
  part1 = part1.decode('utf8').encode('gbk')
  subject = re.search(r'(?<=/>).+', part2).group()
  subject = part1.decode('gbk') + "    " + subject.decode('gbk')
  print( subject.encode('gbk'))

 for i in xrange(num_times - 1):
  url3 = '''
  http://bjapp6.mail.tom.com/cgi/ldapapp?funcid=mails&sid=%s&fid=1&start=%d
  ''' % (sid, (i + 1) * num_per_page )
  response = opener.open(url3)
  data3 = response.read()

  for match in re.finditer(r'(?<="Mbox_Td_Subject"/>).*?(?=/</)', data3):
   index += 1
   part1 = "第%d封郵件" % index
   part2 = " %s" % match.group()
   part1 = part1.decode('utf8').encode('gbk')
   subject = re.search(r'(?<=/>).+', part2).group()
   subject = part1.decode('gbk') + "    " + subject.decode('gbk')
   print( subject.encode('gbk'))