如何在pandas數據幀中讀取mongodb導出的Json

Question

我正在使用以下代碼從mongoDB查詢中導出json：

 querywith open(r'/Month/Applications_test.json', 'w') as f:
for x in dic:
    json.dump(x, f, default=json_util.default)

這很好用，並返回以下json：

{
  "_class": "Application",
  "_id": "123",
  "applicationTimeStamp": {
    "$date": 1541466008000
  },
  "createdDateTime": {
    "$date": 1541466008084
  }
}
{
  "_class": "Application",
  "_id": "124",
  "applicationTimeStamp": {
    "$date": 1540080000000
  },
  "createdDateTime": {
    "$date": 1540080000096
  }
}
{
  "_class": "Application",
  "_id": "125",
  "applicationTimeStamp": {
    "$date": 1540080000000
  },
  "createdDateTime": {
    "$date": 1540080000097
  }
}

我正在使用以下pandas代碼嘗試閱讀它：

data_df = pd.read_json(r'/Month/Applications_test.json', lines = True)

我收到以下錯誤：

ValueError: Unexpected character found when decoding array value (2)

我想要的是一個pandas數據幀：

_class      | _id | applicationTimeStamp | createdDateTime
Application | 123 | 10/07/2018           | 10/07/2018
Application | 124 | 10/07/2018           | 10/07/2018
Application | 125 | 10/07/2018           | 10/07/2018

我怎么能將上面的json讀入熊貓數據幀？

謝謝！

Answer 1

你必須以這種方式使用read_json：

df = pd.read_json(path_or_buf="file_path\json.txt",  typ='frame')

它返回一個數據幀，如：

            _class  _id  applicationTimeStamp  createdDateTime
$date  Application  123         1541466008000    1541466008084

要么：

        _class            ...                      createdDateTime
0  Application            ...             {'$date': 1541466008084}
1  Application            ...             {'$date': 1540080000096}
2  Application            ...             {'$date': 1540000000097}

那么你可以轉換時間戳。