Question

我正在尝试构建JSON布局。我正在从输入文件中读取所有这些记录。该文件中可能有多个具有相同键（Id）的记录。

示例输入文件：

Id,LineNo,Amt,ReceivedDt,FromDt,ToDate,regionId
123545,1,1000.00,2019-02-01T00:00:00,2019-02-01T00:00:00,2019-02-01T00:00:00,WA12
123545,2,200.00,2019-02-01T00:00:00,2019-02-01T00:00:00,2019-02-01T00:00:00,WA12
123545,3,200.00,2019-02-01T00:00:00,2019-02-01T00:00:00,2019-02-01T00:00:00,WA12
123546,1,200.00,2019-02-01T00:00:00,2019-02-01T00:00:00,2019-02-01T00:00:00,WA13
123546,2,200.00,2019-02-01T00:00:00,2019-02-01T00:00:00,2019-02-01T00:00:00,WA13

我的逻辑是以字典格式从文件中读取记录并保持附加直到相同的键（Id）匹配。如果键不再匹配，请删除列表并追加新键，然后将记录与此新键进行比较。在这两者之间，需要存储结果，以免丢失以前处理过的记录。（这是我无法弄清楚的）。

代码：

import json,csv

with open('Test.csv') as f:
    inputfile = csv.DictReader(f)
    output = []
    key =1
    for row in inputfile :

        if len(output)==0:
            output.append(row)

        elif len(output)>0:
            if row['Id']==key:
                output.append(row)

            else:
                del output[:]
                output.append(row)
        key=row['Id']
        data = json.dumps({"data":output}, indent=4)

print(data)

输出： 由于最后一组被删除，只有最后两行会出现。
请建议如何存储这些行。

{
    "data": [
        {
            "ToDate": "2019-02-01T00:00:00",
            "ReceivedDt": "2019-02-01T00:00:00",
            "regionId": "WA13",
            "Id": "123546",
            "LineNo": "1",
            "Amt": "200.00",
            "FromDt": "2019-02-01T00:00:00"
        },
        {
            "ToDate": "2019-02-01T00:00:00",
            "ReceivedDt": "2019-02-01T00:00:00",
            "regionId": "WA13",
            "Id": "123546",
            "LineNo": "2",
            "Amt": "200.00",
            "FromDt": "2019-02-01T00:00:00"
        }
    ]
}

所需的输出：

{
    "data": [
        {
            "ToDate": "2019-02-01T00:00:00",
            "ReceivedDt": "2019-02-01T00:00:00",
            "regionId": "WA12",
            "Id": "123545",
            "LineNo": "1",
            "Amt": "1000.00",
            "FromDt": "2019-02-01T00:00:00"
        },
        {
            "ToDate": "2019-02-01T00:00:00",
            "ReceivedDt": "2019-02-01T00:00:00",
            "regionId": "WA12",
            "Id": "123545",
            "LineNo": "2",
            "Amt": "200.00",
            "FromDt": "2019-02-01T00:00:00"
        },
        {
            "ToDate": "2019-02-01T00:00:00",
            "ReceivedDt": "2019-02-01T00:00:00",
            "regionId": "WA12",
            "Id": "123545",
            "LineNo": "3",
            "Amt": "200.00",
            "FromDt": "2019-02-01T00:00:00"
        }
    ]
},
{
    "data": [
        {
            "ToDate": "2019-02-01T00:00:00",
            "ReceivedDt": "2019-02-01T00:00:00",
            "regionId": "WA13",
            "Id": "123546",
            "LineNo": "1",
            "Amt": "200.00",
            "FromDt": "2019-02-01T00:00:00"
        },
        {
            "ToDate": "2019-02-01T00:00:00",
            "ReceivedDt": "2019-02-01T00:00:00",
            "regionId": "WA13",
            "Id": "123546",
            "LineNo": "2",
            "Amt": "200.00",
            "FromDt": "2019-02-01T00:00:00"
        }
    ]
}

Answer 1

使用>>> type('Ho', (), {'__module__': 'hey'}) <class 'hey.Ho'>：

itertools.groupby

Answer 2

尽管不像使用itertools.groupby那样简洁明了，但这是一种可以手动跟踪具有相同Id的数据组的方法：

import csv
import json

with open('Test.csv') as f:
    output = []
    data = []
    key = None

    for row in csv.DictReader(f):
        if row['Id'] == key:
            data.append(row)
        else:
            if data:
                output.append({"data": data})
                data = []
            data.append(row)
            key = row['Id']

    if data:  # A final group?
        output.append({"data": data})

print('output:\n', json.dumps(output, indent=4))

输出：

output:
 [
    {
        "data": [
            {
                "Id": "123545",
                "LineNo": "1",
                "Amt": "1000.00",
                "ReceivedDt": "2019-02-01T00:00:00",
                "FromDt": "2019-02-01T00:00:00",
                "ToDate": "2019-02-01T00:00:00",
                "regionId": "WA12"
            },
            {
                "Id": "123545",
                "LineNo": "2",
                "Amt": "200.00",
                "ReceivedDt": "2019-02-01T00:00:00",
                "FromDt": "2019-02-01T00:00:00",
                "ToDate": "2019-02-01T00:00:00",
                "regionId": "WA12"
            },
            {
                "Id": "123545",
                "LineNo": "3",
                "Amt": "200.00",
                "ReceivedDt": "2019-02-01T00:00:00",
                "FromDt": "2019-02-01T00:00:00",
                "ToDate": "2019-02-01T00:00:00",
                "regionId": "WA12"
            }
        ]
    },
    {
        "data": [
            {
                "Id": "123546",
                "LineNo": "1",
                "Amt": "200.00",
                "ReceivedDt": "2019-02-01T00:00:00",
                "FromDt": "2019-02-01T00:00:00",
                "ToDate": "2019-02-01T00:00:00",
                "regionId": "WA13"
            },
            {
                "Id": "123546",
                "LineNo": "2",
                "Amt": "200.00",
                "ReceivedDt": "2019-02-01T00:00:00",
                "FromDt": "2019-02-01T00:00:00",
                "ToDate": "2019-02-01T00:00:00",
                "regionId": "WA13"
            }
        ]
    }
]

使用列表生成json布局

2 个答案: