正如在前面的回答中提到的，将json转换为csv的困难在于json文件可以包含嵌套字典，因此是多维数据结构，而csv是2D数据结构。但是，将多维结构转换为csv的一个好方法是使用多个主键连接在一起的csv。

在你的例子中，第一个csv输出的列是“pk”，“model”，“fields”。“pk”和“model”的值很容易获得，但因为“fields”列包含一个字典，它应该是它自己的csv，因为“codename”似乎是主键，你可以使用作为“fields”的输入来完成第一个csv。第二个csv包含来自“fields”列的字典，以codename作为主键，可用于将两个csv绑定在一起。

这是一个解决方案，为您的json文件转换嵌套字典2 csv。

import csv
import json

def readAndWrite(inputFileName, primaryKey=""):
    input = open(inputFileName+".json")
    data = json.load(input)
    input.close()

    header = set()

    if primaryKey != "":
        outputFileName = inputFileName+"-"+primaryKey
        if inputFileName == "data":
            for i in data:
                for j in i["fields"].keys():
                    if j not in header:
                        header.add(j)
    else:
        outputFileName = inputFileName
        for i in data:
            for j in i.keys():
                if j not in header:
                    header.add(j)

    with open(outputFileName+".csv", 'wb') as output_file:
        fieldnames = list(header)
        writer = csv.DictWriter(output_file, fieldnames, delimiter=',', quotechar='"')
        writer.writeheader()
        for x in data:
            row_value = {}
            if primaryKey == "":
                for y in x.keys():
                    yValue = x.get(y)
                    if type(yValue) == int or type(yValue) == bool or type(yValue) == float or type(yValue) == list:
                        row_value[y] = str(yValue).encode('utf8')
                    elif type(yValue) != dict:
                        row_value[y] = yValue.encode('utf8')
                    else:
                        if inputFileName == "data":
                            row_value[y] = yValue["codename"].encode('utf8')
                            readAndWrite(inputFileName, primaryKey="codename")
                writer.writerow(row_value)
            elif primaryKey == "codename":
                for y in x["fields"].keys():
                    yValue = x["fields"].get(y)
                    if type(yValue) == int or type(yValue) == bool or type(yValue) == float or type(yValue) == list:
                        row_value[y] = str(yValue).encode('utf8')
                    elif type(yValue) != dict:
                        row_value[y] = yValue.encode('utf8')
                writer.writerow(row_value)

readAndWrite("data")

使用csv.DictWriter()很容易，详细的实现可以像这样:

def read_json(filename):
    return json.loads(open(filename).read())
def write_csv(data,filename):
    with open(filename, 'w+') as outf:
        writer = csv.DictWriter(outf, data[0].keys())
        writer.writeheader()
        for row in data:
            writer.writerow(row)
# implement
write_csv(read_json('test.json'), 'output.csv')

注意，这假设所有JSON对象都具有相同的字段。

这是一份可能对你有帮助的参考资料。

由于数据看起来是字典格式，因此似乎应该实际使用csv.DictWriter()来实际输出带有适当标题信息的行。这将使转换更容易处理。然后fieldnames参数将正确地设置顺序，而第一行的输出作为标题将允许稍后由csv.DictReader()读取和处理。

例如，Mike Repass使用

output = csv.writer(sys.stdout)

output.writerow(data[0].keys())  # header row

for row in data:
  output.writerow(row.values())

不过，只需将初始设置更改为 输出= csv。DictWriter数据(文件集,字段名= [0]. keys ())

注意，由于字典中元素的顺序没有定义，您可能必须显式地创建字段名条目。一旦你这样做了，writerow就可以工作了。然后写操作就像最初显示的那样工作。

首先，JSON包含嵌套对象，因此通常不能直接转换为CSV。你需要把它改成这样:

{
    "pk": 22,
    "model": "auth.permission",
    "codename": "add_logentry",
    "content_type": 8,
    "name": "Can add log entry"
},
......]

下面是我的代码来生成CSV:

import csv
import json

x = """[
    {
        "pk": 22,
        "model": "auth.permission",
        "fields": {
            "codename": "add_logentry",
            "name": "Can add log entry",
            "content_type": 8
        }
    },
    {
        "pk": 23,
        "model": "auth.permission",
        "fields": {
            "codename": "change_logentry",
            "name": "Can change log entry",
            "content_type": 8
        }
    },
    {
        "pk": 24,
        "model": "auth.permission",
        "fields": {
            "codename": "delete_logentry",
            "name": "Can delete log entry",
            "content_type": 8
        }
    }
]"""

x = json.loads(x)

f = csv.writer(open("test.csv", "wb+"))

# Write CSV Header, If you dont need that, remove this line
f.writerow(["pk", "model", "codename", "name", "content_type"])

for x in x:
    f.writerow([x["pk"],
                x["model"],
                x["fields"]["codename"],
                x["fields"]["name"],
                x["fields"]["content_type"]])

你会得到如下输出:

pk,model,codename,name,content_type
22,auth.permission,add_logentry,Can add log entry,8
23,auth.permission,change_logentry,Can change log entry,8
24,auth.permission,delete_logentry,Can delete log entry,8

使用pandas中的json_normalize:

在名为test.json的文件中使用来自OP的示例数据。 这里使用了Encoding ='utf-8'，但在其他情况下可能不需要。 下面的代码利用了pathlib库。 .open是pathlib的一个方法。 也适用于非windows路径。 使用pandas.to_csv(…)将数据保存为csv文件。

import pandas as pd
# As of Pandas 1.01, json_normalize as pandas.io.json.json_normalize is deprecated and is now exposed in the top-level namespace.
# from pandas.io.json import json_normalize
from pathlib import Path
import json

# set path to file
p = Path(r'c:\some_path_to_file\test.json')

# read json
with p.open('r', encoding='utf-8') as f:
    data = json.loads(f.read())

# create dataframe
df = pd.json_normalize(data)

# dataframe view
 pk            model  fields.codename           fields.name  fields.content_type
 22  auth.permission     add_logentry     Can add log entry                    8
 23  auth.permission  change_logentry  Can change log entry                    8
 24  auth.permission  delete_logentry  Can delete log entry                    8
  4  auth.permission        add_group         Can add group                    2
 10  auth.permission      add_message       Can add message                    4

# save to csv
df.to_csv('test.csv', index=False, encoding='utf-8')

CSV输出:

pk,model,fields.codename,fields.name,fields.content_type
22,auth.permission,add_logentry,Can add log entry,8
23,auth.permission,change_logentry,Can change log entry,8
24,auth.permission,delete_logentry,Can delete log entry,8
4,auth.permission,add_group,Can add group,2
10,auth.permission,add_message,Can add message,4

嵌套更重的JSON对象的资源:

所以答案: 用python平化JSON数组 如何平嵌套的JSON递归，与平坦JSON 如何json_normalize一个列与nan 使用pandas将一列字典拆分为单独的列 有关其他相关问题，请参阅json_normalize标记。

我可能迟到了，但我想，我已经处理过类似的问题。我有一个json文件，看起来像这样

我只想从这些json文件中提取一些键/值。因此，我编写了下面的代码来提取相同的内容。

    """json_to_csv.py
    This script reads n numbers of json files present in a folder and then extract certain data from each file and write in a csv file.
    The folder contains the python script i.e. json_to_csv.py, output.csv and another folder descriptions containing all the json files.
"""

import os
import json
import csv


def get_list_of_json_files():
    """Returns the list of filenames of all the Json files present in the folder
    Parameter
    ---------
    directory : str
        'descriptions' in this case
    Returns
    -------
    list_of_files: list
        List of the filenames of all the json files
    """

    list_of_files = os.listdir('descriptions')  # creates list of all the files in the folder

    return list_of_files


def create_list_from_json(jsonfile):
    """Returns a list of the extracted items from json file in the same order we need it.
    Parameter
    _________
    jsonfile : json
        The json file containing the data
    Returns
    -------
    one_sample_list : list
        The list of the extracted items needed for the final csv
    """

    with open(jsonfile) as f:
        data = json.load(f)

    data_list = []  # create an empty list

    # append the items to the list in the same order.
    data_list.append(data['_id'])
    data_list.append(data['_modelType'])
    data_list.append(data['creator']['_id'])
    data_list.append(data['creator']['name'])
    data_list.append(data['dataset']['_accessLevel'])
    data_list.append(data['dataset']['_id'])
    data_list.append(data['dataset']['description'])
    data_list.append(data['dataset']['name'])
    data_list.append(data['meta']['acquisition']['image_type'])
    data_list.append(data['meta']['acquisition']['pixelsX'])
    data_list.append(data['meta']['acquisition']['pixelsY'])
    data_list.append(data['meta']['clinical']['age_approx'])
    data_list.append(data['meta']['clinical']['benign_malignant'])
    data_list.append(data['meta']['clinical']['diagnosis'])
    data_list.append(data['meta']['clinical']['diagnosis_confirm_type'])
    data_list.append(data['meta']['clinical']['melanocytic'])
    data_list.append(data['meta']['clinical']['sex'])
    data_list.append(data['meta']['unstructured']['diagnosis'])
    # In few json files, the race was not there so using KeyError exception to add '' at the place
    try:
        data_list.append(data['meta']['unstructured']['race'])
    except KeyError:
        data_list.append("")  # will add an empty string in case race is not there.
    data_list.append(data['name'])

    return data_list


def write_csv():
    """Creates the desired csv file
    Parameters
    __________
    list_of_files : file
        The list created by get_list_of_json_files() method
    result.csv : csv
        The csv file containing the header only
    Returns
    _______
    result.csv : csv
        The desired csv file
    """

    list_of_files = get_list_of_json_files()
    for file in list_of_files:
        row = create_list_from_json(f'descriptions/{file}')  # create the row to be added to csv for each file (json-file)
        with open('output.csv', 'a') as c:
            writer = csv.writer(c)
            writer.writerow(row)
        c.close()


if __name__ == '__main__':
    write_csv()

我希望这能有所帮助。有关此代码如何工作的详细信息，请查看这里