我正在尝试使用 pandas 将 json 文件转换为特定格式。我想在某些列上旋转它答案

【问题标题】：I am trying to pivot a json file using pandas to be in a specific format. I want to pivot it on certain columns我正在尝试使用 pandas 将 json 文件转换为特定格式。我想在某些列上旋转它
【发布时间】：2021-12-03 23:55:33
【问题描述】：

所以我尝试在网上搜索以找到一种可以转换以下json的方法。

{
"eTask_ID": "100",
"Organization": "Power",
"BidID": "2.00",
"Project": "IPP - C",
"Forecast%": "67",
"Sponsor": "Jon R",
"IsActive": "1",
"InternalOrder": "null",
"Forecast": "null",
"BidStatus": "null",
"ProjectNotes": "null",
"EstimateTypeCode": "null",
"Start": "null",
"SponsoringDistrict": "null",
"LocationState": "null",
"Finish": "null",
"AreaManager": "null",
"CTG Vendor": "null"
}

到下面的那个。

{
"eTask_ID": "100",
"Organization": "Power",
"BidID": "2.00",
"Project": "IPP - C",
"Attribute":"Forecast%",
"AttrValue":"67",
},
{
"eTask_ID": "100",
"Organization": "Power",
"BidID": "2.00",
"Project": "IPP - C",
"Attribute":"Sponsor",
"AttrValue":"Jon R",
},
{
"eTask_ID": "100",
"Organization": "Power",
"BidID": "2.00",
"Project": "IPP - C",
"Attribute":"IsActive",
"AttrValue":"1",
},
...

现在，如果您看到除前四个之外的所有属性都被转换为 Attribute 和 AttributeValue 并获得它们自己的记录。

我已尝试在网络上搜索解决方案，但仍在尝试寻找解决方案。

如果有人可以请帮忙。

提前谢谢你。

【问题讨论】：

标签： python json pandas dataframe

【解决方案1】：

使用pd.melt:

import json

with open('data.json') as json_data:
    data = json.load(json_data)
    
    out = pd.DataFrame.from_dict(data, orient='index').T \
            .melt(['eTask_ID', 'Organization', 'BidID', 'Project'], 
                  var_name='Attribute', value_name='AttrValue') \
            .to_json(orient='records', indent=4)

输出：

>>> print(out)
[
    {
        "eTask_ID":100,
        "Organization":"Power",
        "BidID":2,
        "Project":"IPP - C",
        "Attribute":"Forecast%",
        "AttrValue":67
    },
    {
        "eTask_ID":100,
        "Organization":"Power",
        "BidID":2,
        "Project":"IPP - C",
        "Attribute":"Sponsor",
        "AttrValue":"Jon R"
    },
    {
        "eTask_ID":100,
        "Organization":"Power",
        "BidID":2,
        "Project":"IPP - C",
        "Attribute":"IsActive",
        "AttrValue":1
    },
    {
        "eTask_ID":100,
        "Organization":"Power",
        "BidID":2,
        "Project":"IPP - C",
        "Attribute":"InternalOrder",
        "AttrValue":"null"
    },
    {
        "eTask_ID":100,
        "Organization":"Power",
        "BidID":2,
        "Project":"IPP - C",
        "Attribute":"Forecast",
        "AttrValue":"null"
    },
    {
        "eTask_ID":100,
        "Organization":"Power",
        "BidID":2,
        "Project":"IPP - C",
        "Attribute":"BidStatus",
        "AttrValue":"null"
    },
    {
        "eTask_ID":100,
        "Organization":"Power",
        "BidID":2,
        "Project":"IPP - C",
        "Attribute":"ProjectNotes",
        "AttrValue":"null"
    },
    {
        "eTask_ID":100,
        "Organization":"Power",
        "BidID":2,
        "Project":"IPP - C",
        "Attribute":"EstimateTypeCode",
        "AttrValue":"null"
    },
    {
        "eTask_ID":100,
        "Organization":"Power",
        "BidID":2,
        "Project":"IPP - C",
        "Attribute":"Start",
        "AttrValue":"null"
    },
    {
        "eTask_ID":100,
        "Organization":"Power",
        "BidID":2,
        "Project":"IPP - C",
        "Attribute":"SponsoringDistrict",
        "AttrValue":"null"
    },
    {
        "eTask_ID":100,
        "Organization":"Power",
        "BidID":2,
        "Project":"IPP - C",
        "Attribute":"LocationState",
        "AttrValue":"null"
    },
    {
        "eTask_ID":100,
        "Organization":"Power",
        "BidID":2,
        "Project":"IPP - C",
        "Attribute":"Finish",
        "AttrValue":"null"
    },
    {
        "eTask_ID":100,
        "Organization":"Power",
        "BidID":2,
        "Project":"IPP - C",
        "Attribute":"AreaManager",
        "AttrValue":"null"
    },
    {
        "eTask_ID":100,
        "Organization":"Power",
        "BidID":2,
        "Project":"IPP - C",
        "Attribute":"CTG Vendor",
        "AttrValue":"null"
    }
]

【讨论】：

非常感谢 Corralien，它工作得非常好。