如何在 Python 中使用多个键解析和排序 JSON

Question

实际使用 JSON 进行培训，试图了解如何使用两个不同的键解析新的 JSON。我想对一些日志进行排序以用于数据可视化。

我的数据 JSON

{
            "productTitle": "Product",
            "apiName": "soapwebservice"
            "statusCode": "200 OK"
},
{
            "productTitle": "Product",
            "apiName": "productionservice",
            "statusCode": "200 OK"
},
{
            "productTitle": "Product",
            "apiName": "firstapi",
            "statusCode": "200 OK"
},          
{
            "productTitle": "Product",
            "apiName": "firstapi",
            "statusCode": "200 OK"
}, 
{
            "productTitle": "Suitability",
            "apiName": "suitability-api",
            "statusCode": "200 OK"
}

预期的输出 JSON：

{
    "Product": 4,
    "api-activity": {
         "soapwebservice": 1,
         "productionservice": 1,
         "firstapi": 2
    }
}
{
    "Suitability": 1,
    "api-activity": {
         "suitability-api": 1,
    }
}

这是我第一个解析和计算第一个键的代码：

import json
from collections import Counter

with open('events1.json') as json_data:
    json_obj = json.load(json_data) # Read the JSON file

    # print(json_obj['calls'][0]['appName']); #Example JSON Extract
    c = Counter(player['productTitle'] for player in json_obj['calls'])

with open('output.json', 'w') as f:
    f.write(json.dumps(c, indent=4)) # Parse and write the file

print("Translation of JSON");

我正在寻找一种获取预期 JSON 的方法，可能带有一个循环，如下所述：parse JSON values by multilevel keys 但是我无法按预期获取 JSON，您知道吗？

Answer 1

使用itertools.groupby() 按所需属性对元素进行分组。第一个按产品标题分组，然后在每个结果组中，再次对元素进行分组，但现在按 api 名称分组。所需的计数器只是每个结果组中元素的数量：

import itertools
import json

def by_product_title(data):
    return data['productTitle']

def by_api_name(data):
    return data['apiName']

json_str = '''
[
    {
        "productTitle": "Product",
        "apiName": "soapwebservice",
        "statusCode": "200 OK"
    },
    {
        "productTitle": "Product",
        "apiName": "productionservice",
        "statusCode": "200 OK"
    },
    ...
]
'''

data = json.loads(json_str)
grouped_by_product_title = itertools.groupby(sorted(data, key=by_product_title), by_product_title)
for product, group in grouped_by_product_title:
    elements = list(group)
    grouped_by_api_name = itertools.groupby(sorted(elements, key=by_api_name), by_api_name)
    api_activity = {key: len(list(val)) for key, val in grouped_by_api_name}
    output = {product: len(elements), 'api-activity': api_activity}
    json_output = json.dumps(output, sort_keys=True, indent=4)
    print(json_output)

输出：

{
    "Product": 4,
    "api-activity": {
        "firstapi": 2,
        "productionservice": 1,
        "soapwebservice": 1
    }
}
{
    "Suitability": 1,
    "api-activity": {
        "suitability-api": 1
    }
}

Answer 2

我确实喜欢这个。看看吧。

import json

json_str = """
[
{
            "productTitle": "Product",
            "apiName": "soapwebservice",
            "statusCode": "200 OK"
},
{
            "productTitle": "Product",
            "apiName": "productionservice",
            "statusCode": "200 OK"
},
{
            "productTitle": "Product",
            "apiName": "firstapi",
            "statusCode": "200 OK"
},
{
            "productTitle": "Product",
            "apiName": "firstapi",
            "statusCode": "200 OK"
},
{
            "productTitle": "Suitability",
            "apiName": "suitability-api",
            "statusCode": "200 OK"
}
]
"""

data_set = json.loads(json_str)
data_list = []
for data in data_set:
    title = data["productTitle"]
    exist = False
    for existing_data in data_list:
        if title in existing_data:
            exist = True
            existing_data[title] += 1
            if data["apiName"] in existing_data["api-activity"]:
                existing_data["api-activity"][data["apiName"]] += 1
            else:
                existing_data["api-activity"][data["apiName"]] = 1
    if not exist:
        new = dict()
        new[data["productTitle"]] = 1
        new["api-activity"] = dict()
        new["api-activity"][data["apiName"]] = 1
        data_list.append(new)

print(json.dumps(data_list))

Answer 3

您的第一个问题是 json.load（和 json.loads）只会从文件中获取一个元素。要解决此问题，您需要将其转换为列表。

[
  {
    "productTitle": "Product",
    "apiName": "soapwebservice",
    "statusCode": "200 OK"
  },
  {
    "productTitle": "Product",
    "apiName": "productionservice",
    "statusCode": "200 OK"
  },
  {
    "productTitle": "Product",
    "apiName": "firstapi",
    "statusCode": "200 OK"
  },
  {
    "productTitle": "Product",
    "apiName": "firstapi",
    "statusCode": "200 OK"
  },
  {
    "productTitle": "Suitability",
    "apiName": "suitability-api",
    "statusCode": "200 OK"
  }
]

您的第二个问题是您试图访问不存在的 json_obj['calls'] 。通过删除访问器，您将获得以下代码：

import json
from collections import Counter

with open('events1.json') as json_data:
    json_obj = json.load(json_data) # Read the JSON file
    c = Counter(player['productTitle'] for player in json_obj)

with open('output.json', 'w') as f:
    f.write(json.dumps(c, indent=4)) # Parse and write the file

print("Translation of JSON");

这会在 output.json 中产生以下输出：

{
  "Product": 4, 
  "Suitability": 1
}