Comparing JSON Responses Between Old and New Services Using Python

This guide shows how to perform data comparison when migrating services by running old and new versions in parallel, initially directing only 10% of traffic to the new service and using the results to verify correctness before a complete switch.

First, install the required dependency:

pip3 install json_tools

For JSON responses, the following Python script reads a list of parameters from a text file, calls the old and new APIs, retrieves their JSON payloads, and uses json_tools.diff to highlight differences:

# encoding:utf-8
'''两个json对比
通过搜索replace查找哪些不一致
'''
import json
import requests
import json_tools
from urllib import parse
import sys
from conf import hosts, apis

#读取txt里的参数作为变量，因为数据对比是尽可能全的包含向上已经存在的数据
resultpath = '../参数.txt'
lines = []
with open(resultpath, 'r') as file_to_read:
    while True:
        line = file_to_read.readline()
        if not line:
            break
        line = line.strip('
')
        lines.append(line)

def test_json():
    for i in lines:
        url_old = hosts["test"] + apis["API_old"]["path"] + i
        res_old = requests.get(url_old, cookies=graycookei)
        query_old = res_old.json()
        url_new = hosts["test"] + apis["API_new"]["path"] + i
        res_new = requests.get(url_new, cookies=graycookei)
        query_new = res_new.json()
        contrast1 = json_tools.diff(query_old["data"]["expressInfo"], query_new["data"]["expressInfo"]))
        # 打印出两个接口返回的json不一致的数据
        print(contrast1, url_new)

test_query_old()

When the service returns data as a list, the script below demonstrates how to find common elements and identify items that are missing in either list using list comprehensions:

list1 = [1, 2.3, 3, 4, 5]
list2 = [2, 3.3, 4, 5]

c = [x for x in list1 if x in list2]
d = [y for y in (list1 + list2) if y not in c]

print(1, c)
print("不一样的数据或缺失的数据", d)

For quick visual comparison of two JSON objects, an online tool such as https://www.sojson.com/compare.html can be used.