将csv转换为json多文档？

Question

下面有两个使用Python的要求：

1. 将csv转换为多文档json.
2. 忽略“或空对象.

”。

同时提到了代码和csv。目前，我只获得json数组对象，但需要在多文档json中创建。

我的csv

_id,riderDetails.0.category,riderDetails.0.code,riderDetails.1.category,riderDetails.1.code
1111,re,remg,er,error
2111,we,were,ty,

码

import csv
import json
def make_record(row):
    return  { 
    "_id" : row["_id"], 
    "riderDetails" : [
        {            
            "category" : row["riderDetails.0.category"], 
            "code" : row["riderDetails.0.code"],
        },
        {            
            "category" : row["riderDetails.1.category"] ,
            "code" : row["riderDetails.1.code"],
        }
    ]
}  


with open('N:/Exide/Mongo/rr22.csv', 'r', newline='') as csvfile:
     reader = csv.DictReader(csvfile, delimiter=',')

     with open('N:/Exide/Mongo/mm22.json', 'w') as jsonfile:       
       out = json.dumps([make_record(row) for row in reader])
       jsonfile.write(out)

码输出

[{
    "_id": "1111",
    "riderDetails": [
        {
            "category": "re",
            "code": "remg"
        },
        {
            "category": "er",
            "code": "error"
        }
    ]
},
{
    "_id": "2111",
    "riderDetails": [
        {
            "category": "we",
            "code": "were"
        },
        {
            "category": "",
            "code": ""
        }
    ]
}]

预期输出

{
    "_id": "1111",
    "riderDetails": [
        {
            "category": "re",
            "code": "remg"
        },
        {
            "category": "er",
            "code": "error"
        }
    ]
}
{
    "_id": "2111",
    "riderDetails": [
        {
            "category": "we",
            "code": "were"
        }
    ]
}

有人能帮我实现预期的产出吗？

Answer 1

问题中的my 文件中的数据不会产生所显示的输出，但这可能是由于一个小的投递错误造成的，所以我将忽略它。

还请注意，您正在生成的文件不是一个严格有效的JSON格式文件--也许这就是术语“多文档json"…的意思

无论如何，您可以通过修改make_record()函数来完成所需的任务，这样它就可以“清理”记录，并在它返回之前删除任何空/缺失的值。

这分两步完成。

values.

• Lastly，

1. 首先从riderDetails中的每个细节开始，然后删除所有有空riderDetails的键，再次遍历每个riderDetails并删除完全为空的任何细节(因为第一步删除了所有内容，或者在正在读取的csv文件中没有提供任何内容)。

import csv
import json

csv_inp = 'rr22.csv'
json_outp = 'mm22.json'

def make_record(row):
    # Reformat data is row.
    record = {
        "_id": row["_id"],
        "riderDetails": [
            {
                "category": row["riderDetails.0.category"],
                "code": row["riderDetails.0.code"],
            },
            {
                "category": row["riderDetails.1.category"],
                "code": row["riderDetails.1.code"],
            }
        ]
    }

    # Remove empty values from each riderDetail.
    record['riderDetails'] = [{key: value for key, value in riderDetail.items() if value}
                                for riderDetail in record['riderDetails']]

    # Remove completely empty riderDetails.
    record['riderDetails'] = [riderDetail for riderDetail in record['riderDetails']
                                if riderDetail]
    return record

with open(csv_inp, 'r', newline='') as csvfile, \
     open(json_outp, 'w') as jsonfile:

    for row in csv.DictReader(csvfile, delimiter=','):
        jsonfile.write(json.dumps(make_record(row), indent=4) + '\n')
#        jsonfile.write(json.dumps(make_record(row)) + '\n')

Answer 2

使用球体

import glob, os

pt = 'N:/Exide/Mongo/*.csv'

for file in glob.glob(pt):
    get_name = file.split("/")[-1].replace(".csv",".json")
    with open(file , 'r', newline='') as csvfile:
        reader = csv.DictReader(csvfile, delimiter=',')
        out = [make_record(row) for row in reader]

    saving_path = os.path.join('N:/Exide/Mongo/',get_name)
    with open(saving_path , 'w') as jsonfile:       
        json.dump(out , jsonfile)

你得到{}，{}因为你把字典的列表写进文件