1

我正在尝试在 Azure sql 中使用 OPENJSON 解析 json 文件，但我想在我的新表中进行特定映射。我有一个 json 文件，它看起来像这样，有很多 y0_1、y0_2、....y0_380。

    {"proto": "01",
     "type": "A",

 "description": "heartbeat",
 "geometry": {"y0_1": {"tag": "Normal",
   "probability": 0.40,
   "x": 39,
   "y": 13},
  "y0_2": {"tag": "category_3", "probability": 0.8, "x": 48, "y": 13},
  "y0_3": {"tag": "Normal", "probability": 0.9, "x": 27, "y": 10},
"Test": {"proba": 0.65}}}

我想解析 json 并有一个具有这种格式的表：我尝试使用 Openjson 格式解析 json，但我不知道如何像链接中的图片一样在同一列中获取所有 y 和 x

SELECT * from OPENROWSET (
        BULK 'sample.json',
        DATA_SOURCE ='dataset',
        FORMATFILE_DATA_SOURCE = 'myblobstorage',
        SINGLE_CLOB
        ) AS J
        CROSS APPLY OPENJSON (BulkColumn)
        WITH (
            proto nvarchar(100) N'$.proto',
            description nvarchar(100) N'$.description',
        
        )

所需的表输出

4

1 回答 1

1

使用OPENJSON,AS JSON数据类型和CROSS APPLY钻取 JSON 的不同层。我创建了一个简化的示例来展示该技术：

DECLARE @json NVARCHAR(MAX) = '{
  "proto": "01",
  "type": "A",
  "description": "heartbeat",
  "geometry": {
    "y0_1": {
      "tag": "Normal",
      "probability": 0.40,
      "x": 39,
      "y": 13
    },
    "y0_2": {
      "tag": "category_3",
      "probability": 0.8,
      "x": 48,
      "y": 13
    },
    "y0_3": {
      "tag": "Normal",
      "probability": 0.9,
      "x": 27,
      "y": 10
    },
    "Test": {
      "proba": 0.65
    }
  }
}'


SELECT
    j.proto,
    j.[type],
    j.[description],
    JSON_VALUE ( g.[value], '$.tag' ) AS tag,
    JSON_VALUE ( g.[value], '$.probability' ) AS probability,
    JSON_VALUE ( g.[value], '$.x' ) AS x,
    JSON_VALUE ( g.[value], '$.y' ) AS y

FROM OPENJSON( @json, '$' )
WITH (
    proto VARCHAR(50),
    [type] VARCHAR(50),
    [description] VARCHAR(50),
    [geometry] NVARCHAR(MAX) AS JSON
    ) j
    CROSS APPLY OPENJSON ( [geometry] ) g

我的结果：

我的结果

于 2020-08-11T19:43:04.137 回答