如何用Python将CSV转换为含数组类型字段的JSON

现有一段Python代码可将CSV文件转换为JSON格式，但当前转换结果中amounts与quantity为独立字段。需要将amounts字段转换为包含quantity和对应数值的数组结构，以下是具体修改方案：

当前代码 #

import csv
import json

csvfile = open('csv/sample1.csv', encoding='utf-8-sig')
jsonfile = open('output/file.json', 'w')

fieldnames = ("amounts","quantity","product","online_from","online_to","price_info")
reader = csv.DictReader( csvfile, fieldnames)
for row in reader:
    json.dump(row, jsonfile)
    jsonfile.write('\n')

当前转换结果 #

{"amounts": "100", "quantity": "1", "product": "1234567890", "online_from": "2022-08-08T09:00:26.000Z", "online_to": null, "price_info": null}
{"amounts": "200", "quantity": "1", "product": "0987654321", "online_from": "2022-08-08T09:00:26.000Z", "online_to": null, "price_info": null}

期望转换结果 #

{"amounts": [{"quantity": "1", "value": "100"}], "product": "1234567890", "online_from": "2022-08-08T09:00:26.000Z", "online_to": null, "price_info": null}
{"amounts": [{"quantity": "1", "value": "200"}], "product": "0987654321", "online_from": "2022-08-08T09:00:26.000Z", "online_to": null, "price_info": null}

注：原期望结果中第二条的value值应为200，此处已修正符合样本数据逻辑

样本CSV数据 #

100,1,1234567890,2022-08-08T09:00:26.000Z

200,1,0987654321,2022-08-08T09:00:26.000Z

修改后的代码 #

import csv
import json

with open('csv/sample1.csv', encoding='utf-8-sig') as csvfile, open('output/file.json', 'w') as jsonfile:
    fieldnames = ("amounts","quantity","product","online_from","online_to","price_info")
    reader = csv.DictReader(csvfile, fieldnames)
    for row in reader:
        # 跳过空行（处理样本中的空行）
        if not any(row.values()):
            continue
        # 重构amounts字段为数组结构
        row['amounts'] = [{'quantity': row['quantity'], 'value': row['amounts']}]
        # 删除原quantity字段
        del row['quantity']
        # 写入JSON文件
        json.dump(row, jsonfile)
        jsonfile.write('\n')

修改说明 #

1. 使用with语句管理文件，自动处理文件关闭，避免资源泄漏
2. 增加空行判断，跳过CSV中的空行数据
3. 重构amounts字段：将原amounts的值作为value，原quantity的值保留，组合成字典后放入数组
4. 删除原quantity字段，避免结果中出现冗余字段

内容的提问来源于stack exchange，提问作者DM Equinox