如何从 csv 转换为嵌套的 json 字典（初学者）答案

【问题标题】：How to convert from csv to nested json dictionary ( A beginner )如何从 csv 转换为嵌套的 json 字典（初学者）
【发布时间】：2021-10-23 17:26:02
【问题描述】：

我的结果：

[
    {
        "FIRST NAME": "JOHN",
        "PRY SCHOOL": "OLIVETTE",
        "HIGH SCHOOL": "BAPTIST",
        "VEHICLEMAKE ": "TOYOTA",
        "VEHICLE COL": "BLACK",
        "TV MAKE": "SAMSUNG"
    },
    {
        "FIRST NAME": "KOFI",
        "PRY SCHOOL": "ACADAMY", 
        "HIGH SCHOOL": "MAYFLOWER",
        "VEHICLEMAKE ": "HONDA",
        "VEHICLE COL": "YELLOW",
        "TV MAKE": "TECHWOOD"
    },
    {
        "FIRST NAME": "BISI",
        "PRY SCHOOL": "IGBOBI",
        "HIGH SCHOOL": "ANGUS",
        "VEHICLEMAKE ": "HYUNDAI",
        "VEHICLE COL": "BLUE",
        "TV MAKE": "THERMOC"
    }
]

预期结果：

[
    {
        "FIRST NAME": "JOHN",
        "SCHOOL": {
            "primary": "OLIVETTE",
            "HIGH SCHOOL": "BAPTIST"
        },
        "VEHICLE": {
            "MAKE": "TOYOTA",
            "COL": "BLACK"
        },
        "TV MAKE": "SAMSUNG"
    },
    {
        "FIRST NAME": "KOFI",
        "SCHOOL": {
            "primary": "ACADAMY",
            "HIGH SCHOOL": "MAYFLOWER"
        },
        "VEHICLE": {
            "MAKE": "HONDA",
            "COL": "YELLOW"
        },
        "TV MAKE": "TECHWOOD"
    },
    {
        "FIRST NAME": "BISI",
        "SCHOOL": {
            "primary": "IGBOBI",
            "HIGH SCHOOL": "ANGUS"
        },
        "VEHICLE": {
            "MAKE": "HYUNDAI",
            "COL": "BLUE"
        },
        "TV MAKE": "THERMO"
    }
]

我的代码：

import csv
import json

filenames = 'csvfilepath.csv'
my_dic = []

with open(filenames, encoding='utf-8') as csv_file:
     csv_reader = csv.DictReader(csv_file)

     for row in csv_reader:
         my_dic.append(row)

with open('jasonfilepath.json', 'w', encoding='utf-8') as file_object:
    json.dump(my_dic, file_object ,indent = 4)

我的数据：

firstname  PRY SCHOOL HIGH SCHOOL VEHICLEMAKE VEHICLECOL  TVMAKE
JOHN       OLIVETTE   BAPTIST     TOYOTA      BLACK       SAMSUNG
KOFI       ACADAMY    MAYFLOWER    HONDA      YELLOW      TECHWOOD
BISI       IGBOBI     ANGUS        HYUNDAI    BLUE         THERMO

注意：行数超过 1000 行

我想确保学校（包括小学和中学）和车辆（品牌和颜色）的嵌套结构

【问题讨论】：

到目前为止你尝试了什么？
您的输入 CSV 数据文件选项卡是否分开？

标签： python json dictionary

【解决方案1】：

DictReader 无法读取嵌套数据，需要手动构造所需结构的字典。对于这种情况，我会使用简单的csv.reader。

代码：

import csv
import json

with open(r"csvfilepath.csv", newline="") as inp_f, \
        open(r"jsonfilepath.json", "w") as out_f:
    reader = csv.reader(inp_f, delimiter="\t")
    next(reader)  # skip header
    my_dic = []
    for row in reader:
        if len(row) >= 6:  # skip rows which missing columns
            my_dic.append({
                "FIRST NAME": row[0],
                "SCHOOL": {
                    "primary": row[1],
                    "HIGH SCHOOL": row[2]
                },
                "VEHICLE": {
                    "MAKE": row[3],
                    "COL": row[4]
                },
                "TV MAKE": row[5]
            })
    if my_dic:  # if my_dic is not empty
        json.dump(my_dic, out_f, indent=4)

【讨论】：

请我解释一下为什么当 len(row)>=6 被删除时此代码将不起作用我试图删除它并且代码失败。我需要解释以便更好地了解正在发生的事情。
@AbdRaheem，它检查行中是否至少有 6 列。

【解决方案2】：

类似于以下内容（您确定“TV MAKE”的位置吗？）

import json
data = [ { "FIRST NAME": "JOHN", "PRY SCHOOL": "OLIVETTE", "HIGH SCHOOL": "BAPTIST", "VEHICLEMAKE ": "TOYOTA", "VEHICLE COL": "BLACK", "TV MAKE": "SAMSUNG" }, { "FIRST NAME": "KOFI", "PRY SCHOOL": "ACADAMY", "HIGH SCHOOL": "MAYFLOWER", "VEHICLEMAKE ": "HONDA", "VEHICLE COL": "YELLOW", "TV MAKE": "TECHWOOD" }, { "FIRST NAME": "BISI", "PRY SCHOOL": "IGBOBI", "HIGH SCHOOL": "ANGUS", "VEHICLEMAKE ": "HYUNDAI", "VEHICLE COL": "BLUE", "TV MAKE": "THERMOC" } ]


new = [{'SCHOOL':{'PRIMARY':d['PRY SCHOOL'],'HIGH SCHOOL':d['HIGH SCHOOL']},'VEHICLE':{'MAKE':d['VEHICLEMAKE '],'COL':d['VEHICLE COL']},'FIRST NAME':d['FIRST NAME'],'TV MAKE':d['TV MAKE']} for d in data]

print(json.dumps(new,indent=4))

输出

[
    {
        "SCHOOL": {
            "PRIMARY": "OLIVETTE",
            "HIGH SCHOOL": "BAPTIST"
        },
        "VEHICLE": {
            "MAKE": "TOYOTA",
            "COL": "BLACK"
        },
        "FIRST NAME": "JOHN",
        "TV MAKE": "SAMSUNG"
    },
    {
        "SCHOOL": {
            "PRIMARY": "ACADAMY",
            "HIGH SCHOOL": "MAYFLOWER"
        },
        "VEHICLE": {
            "MAKE": "HONDA",
            "COL": "YELLOW"
        },
        "FIRST NAME": "KOFI",
        "TV MAKE": "TECHWOOD"
    },
    {
        "SCHOOL": {
            "PRIMARY": "IGBOBI",
            "HIGH SCHOOL": "ANGUS"
        },
        "VEHICLE": {
            "MAKE": "HYUNDAI",
            "COL": "BLUE"
        },
        "FIRST NAME": "BISI",
        "TV MAKE": "THERMOC"
    }
]

【讨论】：

不，谢谢。我已经编辑了我的问题。我想在学校（包括小学和中学）和车辆（包括车辆品牌和颜色）有一个嵌套结构，行数超过1000行，所以我们不能使用上面的方法。
电视品牌和名字没有嵌套。
@AbdRaheem 查看输出。这是您正在寻找的确切输出。你能解释一下缺少什么吗？

【解决方案3】：

假设制表符分隔输入文件，您可以在读取输入文件时通过稍微修改逻辑来构建嵌套：

with open(filenames, encoding='utf-8') as csv_file:
     csv_reader = csv.DictReader(csv_file, delimiter="\t")
     for row in csv_reader:
         my_dic.append({"FIRST NAME": row['firstname'],
                        "SCHOOL" : { "PRIMARY": row['PRY SCHOOL'],
                                     "HIGH SCHOOL" : row['HIGH SCHOOL']},
                        "VEHICLE" : { "MAKE": row['VEHICLEMAKE'],
                                      "COL": row['VEHICLECOL']},
                        "TV MAKE" : row["TVMAKE"]})

【讨论】：