mitmprxoy工具+python实现高定制抓包

1.mitmproxy扩展机制

参考官网：

mitmproxy addons example： https://docs.mitmproxy.org/stable/addons-examples/#example-complexhar_dumppy 

2.抓包-map local

以雪球app为例

1）编写脚本,mock2.json中自选股列表数据加倍

mitmproxydemo.py

from mitmproxy import http


def request(flow: http.HTTPFlow) -> None:
    if "quote.json" in flow.request.pretty_url and "x=" in flow.request.pretty_url:
        with open("/Users/zhaitiantian3/Public/mock2.json") as f:
            flow.response = http.HTTPResponse.make(200, f.read(), {"Content-Type": "application/json"})

2）手机app连接代理，安装证书：mit.it　

3）运行脚本

mitmdump -s mitmproxydemo.py

刷新自选股列表数据刷新，数据加倍；但是如果是ios手机，我这里使用代理，有的app就会报错

<< Cannot establish TLS with 182.92.251.113:443 (sni: None): TlsException('Cannot validate certificate hostname without SNI')　　

因此查找解决方案：

 

使用命令：

mitmdump -s -k mitmproxydemo.py

mock后的数据：

3.抓包-rewrite

1）编写python脚本

rewriter.py

import json


def response(flow):
    if "quote.json" in flow.request.pretty_url and "x=" in flow.request.pretty_url:
        data = json.loads(flow.response.content)
        data['data']['items'][0]['quote']['name'] = data['data']['items'][0]['quote']['name'] + "test"
        flow.response.text = json.dumps(data)

2）手机app连接代理，安装证书：mit.it　

3）运行脚本

mitmdump -s rewriter.py

4）数据展示如下：

4.自动生成等价类数据

编写脚本：

import json

from mitmproxy import http

url_index = dict()
arrays = [-5, -3, -1, 0, 1, 3, 5, 100]


def response(flow: http.HTTPFlow) -> None:
    # if "Content-Type" in flow.response.headers.keys() and \
    #         "json" in flow.response.headers['Content-Type']:
    if "quote.json" in flow.request.pretty_url and "x=" in flow.request.pretty_url:
        url = flow.request.url.split('.json')[0]
        if url not in url_index.keys():
            url_index[url] = 0
        else:
            url_index[url] += 1

        # 去等价类中的某一个，根据访问次数循环使用
        seed = url_index[url] % len(arrays)
        print(seed)
        data = json.loads(flow.response.text)
        # 对数据进行批量修改
        data_new = json_travel(data, num=arrays[seed])
        json_new = json.dumps(data_new, indent=2)
        flow.response.text = json_new


def json_travel(data, array=None, text=1, num=1):
    data_new = None
    # 如果是词典，对词典进行遍历
    if isinstance(data, dict):
        data_new = dict()
        for k, v in data.items():
            data_new[k] = json_travel(v, array, text, num)

    # 如果是列表，对列表的每一项进行遍历
    elif isinstance(data, list):
        data_new = list()
        for item in data:
            item_new = json_travel(item, array, text, num)
            if array is None:
                data_new.append(item_new)
            elif len(data_new) < array:
                data_new.append(item_new)
            else:
                pass
    # 如果是字符串
    elif isinstance(data, str):
        data_new = data * text
    # 如果是int或者float这样的数字
    elif isinstance(data, int) or isinstance(data, float):
        # 对数字进行一个乘积计算
        data_new = data * num
    # 其他数据类型保持原样
    else:
        data_new = data
    return data_new


def test_json_travel():
    with open("demo.json") as f:
        data = json.load(f)
        print(json_travel(data, array=0))
        print(json_travel(data, text=5))
        print(json_travel(data, num=5))

2）手机app连接代理，安装证书：mit.it　

3）运行脚本

mitmdump -s tstcase.py

根据访问次数循环乘当前价格，可以使用该方式mock出其他前端的逻辑