Python使用Mitmproxy进行HTTP抓包与流量监控实战

pip install mitmproxy

"""
flow.request.scheme 请求协议
flow.request.host 请求host
flow.request.url  请求URL链接
flow.request.method 请求方法
flow.request.query 请求URL查询参数
flow.request.path 请求URL https://www.baidu.com/ 
flow.request.path_components #请求URL不包含域名的元祖 ('project', 'classify', 'list')
flow.request.urlencoded_form 请求POST数据
flow.response.status_code HTTP响应状态码
flow.response.headers HTTP响应头信息
flow.response.get_text HTTP响应内容

"""

class Counter:
 def __init__(self):
  self.result = {} # 存接口请求和返回信息
  # url filter 去掉
  self.url_filter = ['baidu.com','qq.com','360']
  # url screen 仅访问
  self.url_race = ['10.162.16.39:8091']
  # http static extension
  self.static_ext = ['js', 'css', 'ico', 'jpg', 'png', 'gif', 'jpeg', 'bmp','xml']
  # http Content-Type
  self.static_files = ['text/css','image/jpeg', 'image/gif','image/png','text/html','application/octet-stream','application/x-protobuf']
  # http Content-Type media resource files type
  self.media_types = ['image', 'video', 'audio']

 def parser_data(self,query,data = {}):
  for key, value in query.items():
   data[key] = value
  return data

 def get_extension(self, url_tup):
  if not url_tup:
   return ''
  else:
   end_path = url_tup[-1]
   split_ext = end_path.split('.')  #1148e88a9d97.jpg #list
   return '' if not split_ext or len(split_ext) == 1 else split_ext[-1]

 # 拒绝连接
 def http_connect(self, flow: mitmproxy.http.HTTPFlow):
  for i in self.url_filter: #过滤url
   if i in flow.request.host:
    flow.response = http.HTTPResponse.make(404)

 #存在筛选就返回true拦截，Flase通过
 def capture_pass(self,request,response):
  if self.url_race:
   if request.host not in self.url_race: #筛选url
    return True
  url_tup = request.path_components #获取url的tup
  extension = self.get_extension(url_tup)
  if extension in self.static_ext: #判断后缀
   return True
  try:
   content_type = response.headers['Content-Type'].split(';')[0]
   if not content_type:
    return False
   elif content_type in self.static_files: #判断Content-Type
    return True
   else:
    http_mime_type = content_type.split('/')[0]
    if http_mime_type in self.media_types: #判断Content-Type的files type
     return True
    else:
     return False
  except Exception:
   return False



 def request(self, flow: mitmproxy.http.HTTPFlow):
  request = flow.request
  # 修改请求头
  # request.headers["shuzf"] = "shuzf"
  # # 修改get参数
  # if "shuzf" in flow.request.query.keys():
  #  request.query.set_all("shuzf", ["舒志福"])
  # # 修改post参数
  # if "shuzf" in flow.request.urlencoded_form.keys():
  #  request.urlencoded_form.set_all('shuzf', '舒志福')
  scheme = request.scheme
  domain = request.host
  self.result['url'] = parse.unquote(request.url) # url解码
  self.result['method'] = request.method
  self.result['request_headers'] = {}
  for item in request.headers:
   self.result['request_headers'][item] = request.headers[item]
  self.result['get_data'] = self.parser_data(request.query) # 将表单转字典
  self.result['post_data'] = self.parser_data(request.urlencoded_form) # 将表单转字典

 def response(self, flow: mitmproxy.http.HTTPFlow):
  request = flow.request
  response = flow.response
  # # 修改返回头
  # response.headers["shuzf"] = "shuzf"
  # # 修改返回体
  # text = response.text
  # text = text.replace("shuzf", "舒志福")
  # flow.response.set_text(text)
  if not self.capture_pass(request,response):
   print(request.url)
   self.result['status_code '] = response.status_code
   self.result['response_headers'] = {}
   for item in response.headers:
    self.result['response_headers'][item] = response.headers[item]
   # HTTPResponse内部使用了iso-8859-1编码，先进行解码为Unicode再进行utf-8编码 response.text.encode("iso-8859-1").decode("utf-8")
   self.result['response_content'] = response.text 
   # 添加result至数据库
   new_url = Proxy(url=self.result['url'],res=self.result['response_content'], content=json.dumps(self.result))

   session.add(new_url)
   session.commit()

   # 关闭session:
   # session.close()

addons = [Counter()] # 实例类