指熊猫中的父属性

Question

这是我的json

{
    "fInstructions": [
        {
            "id": 155,
            "type":"finstruction",
            "ref": "/spm/finstruction/155",
            "iLineItem":[
                {
                    "id": 156,
                    "type":"ilineitem",
                    "ref": "/spm/ilineitem/156",
                    "creationDate": "2018-03-09",
                    "dueDate":"2018-02-01",
                    "effectiveDate":"2018-03-09",
                    "frequency":"01",
                    "coveredPeriodFrom":"2018-02-28",
                    "coveredPeriodTo":"2018-02-28",
                    "statusCode":"PRO",
                    "amount": 6
                },
                {
                    "id": 157,
                    "type":"ilineitem",
                    "ref": "/spm/ilineitem/157",
                    "creationDate": "2018-03-09",
                    "dueDate":"2018-02-01",
                    "effectiveDate":"2018-03-09",
                    "frequency":"01",
                    "coveredPeriodFrom":"2018-03-01",
                    "coveredPeriodTo":"2018-03-31",
                    "statusCode":"PRO",
                    "amount": 192
                }
            ]
        }
    ]
}

如果我这样做：

json_normalize(data['fInstructions'], record_path=['iLineItem'])

我按预期得到两行所有 ILI。 但是，我还想在结果集中键入父属性 id。 为此，我尝试：

json_normalize(df_data_1['fInstructions'], record_path=['iLineItem'], meta=['id', 'type'])

但后来我得到： 值错误：元数据名称 ID 冲突，需要区分前缀

所以我尝试：

json_normalize(df_data_1['fInstructions'], record_path=['iLineItem'], meta=['fInstructions.id'])

这给了我：

KeyError: "Try running with errors='ignore' as key 'fInstructions.id' is not always present"

Answer 1

答案是：

json_normalize(df_data_1['fInstructions'], record_path=['iLineItem'], meta='id', record_prefix='ils.')