路径包含有趣字符时的 TSQL OPENJSON
Posted
技术标签:
【中文标题】路径包含有趣字符时的 TSQL OPENJSON【英文标题】：TSQL OPENJSON when path has funky charactors 【发布时间】：2020-07-01 06:53:32 【问题描述】：我的 JSON 金融时间序列数据来自 alphavantage
我怀疑路径标签中没有空格和句点 (.) 会很好。我设法让 TSQL 在没有解析器大惊小怪但没有路径节点数据的情况下运行。特殊字符是原因吗？
DECLARE @JSON NVARCHAR(MAX)
SELECT @JSON =
'
"Meta Data":
"1. Information": "Daily Prices (open, high, low, close) and Volumes",
"2. Symbol": "XLK",
"3. Last Refreshed": "2020-06-30",
"4. Output Size": "Full size",
"5. Time Zone": "US/Eastern"
,
"Time Series (Daily)":
"2020-06-30":
"1. open": "102.6600",
"2. high": "104.9100",
"3. low": "102.5200",
"4. close": "104.4900",
"5. volume": "8061852"
,
"2020-06-29":
"1. open": "101.5300",
"2. high": "102.6600",
"3. low": "100.3100",
"4. close": "102.6500",
"5. volume": "9271548"
,
"2020-06-26":
"1. open": "103.3800",
"2. high": "103.5300",
"3. low": "101.2300",
"4. close": "101.5400",
"5. volume": "22135104"
,
"2020-06-25":
"1. open": "102.4200",
"2. high": "103.6900",
"3. low": "101.3100",
"4. close": "103.5800",
"5. volume": "10265454"
,
"2020-06-24":
"1. open": "104.2000",
"2. high": "104.7800",
"3. low": "101.7100",
"4. close": "102.2700",
"5. volume": "12710857"
,
"2020-06-23":
"1. open": "104.8100",
"2. high": "105.8300",
"3. low": "104.4800",
"4. close": "104.6300",
"5. volume": "9508806"
,
"2020-06-22":
"1. open": "102.0900",
"2. high": "103.9700",
"3. low": "101.9300",
"4. close": "103.8800",
"5. volume": "7365608"
,
"2020-06-19":
"1. open": "104.1900",
"2. high": "104.3100",
"3. low": "101.7500",
"4. close": "102.2400",
"5. volume": "15638477"
,
"2020-06-18":
"1. open": "102.3000",
"2. high": "103.0600",
"3. low": "102.0800",
"4. close": "102.9500",
"5. volume": "6853710"
,
"2020-06-17":
"1. open": "103.0900",
"2. high": "103.4600",
"3. low": "102.3200",
"4. close": "102.4900",
"5. volume": "8655904"
,
"2020-06-16":
"1. open": "102.7700",
"2. high": "103.3900",
"3. low": "100.9200",
"4. close": "102.5100",
"5. volume": "13967902"
,
"2020-06-15":
"1. open": "97.6000",
"2. high": "100.8200",
"3. low": "97.2800",
"4. close": "100.3700",
"5. volume": "14961914"
,
"2020-06-12":
"1. open": "100.7700",
"2. high": "101.2700",
"3. low": "97.6800",
"4. close": "99.4100",
"5. volume": "21642002"
,
"2020-06-11":
"1. open": "102.3000",
"2. high": "102.4500",
"3. low": "98.1100",
"4. close": "98.1400",
"5. volume": "19867498"
,
"2020-06-10":
"1. open": "103.1400",
"2. high": "104.8900",
"3. low": "102.9500",
"4. close": "104.1100",
"5. volume": "12177473"
,
"2020-06-09":
"1. open": "101.4100",
"2. high": "102.9000",
"3. low": "101.3200",
"4. close": "102.4000",
"5. volume": "7352624"
,
"2020-06-08":
"1. open": "101.2300",
"2. high": "101.9600",
"3. low": "100.3000",
"4. close": "101.9200",
"5. volume": "8064775"
,
"2020-06-05":
"1. open": "99.5100",
"2. high": "101.7000",
"3. low": "99.3400",
"4. close": "101.4100",
"5. volume": "10374918"
'
SELECT
*
FROM OPENJSON(@JSON,'$."Time Series (Daily)"')
WITH(
series_open VARCHAR(18) '$."1. open"',
searies_high VARCHAR(128) '$."2. high"',
series_close VARCHAR(128) '$."4. close"',
searies_volumn VARCHAR(28) '$."5. volume"'
)
很高兴拥有： 为整个集合重复的元数据 路径中 5 个子节点中每个子节点的重复日期。
感谢专家！
【问题讨论】：
【参考方案1】：以下内容足以让您自己解决这个问题：
您的 JSON（减少到仅一些节点）：
DECLARE @JSON NVARCHAR(MAX)=
N'
"Meta Data":
"1. Information": "Daily Prices (open, high, low, close) and Volumes",
"2. Symbol": "XLK",
"3. Last Refreshed": "2020-06-30",
"4. Output Size": "Full size",
"5. Time Zone": "US/Eastern"
,
"Time Series (Daily)":
"2020-06-30":
"1. open": "102.6600",
"2. high": "104.9100",
"3. low": "102.5200",
"4. close": "104.4900",
"5. volume": "8061852"
,
"2020-06-29":
"1. open": "101.5300",
"2. high": "102.6600",
"3. low": "100.3100",
"4. close": "102.6500",
"5. volume": "9271548"
,
"2020-06-26":
"1. open": "103.3800",
"2. high": "103.5300",
"3. low": "101.2300",
"4. close": "101.5400",
"5. volume": "22135104"
'
--查询
SELECT JSON_VALUE(A.[Meta Data],'$."1. Information"') AS Meta_Information
,JSON_VALUE(A.[Meta Data],'$."2. Symbol"') AS Meta_Symbol
,JSON_VALUE(A.[Meta Data],'$."3. Last Refreshed"') AS Meta_LastRefreshed
,JSON_VALUE(A.[Meta Data],'$."4. Output Size"') AS Meta_OutputSize
,JSON_VALUE(A.[Meta Data],'$."5. Time Zone"') AS Meta_TimeZone
,B.[key] AS Data_Date
,C.*
FROM OPENJSON(@JSON) WITH([Meta Data] NVARCHAR(MAX) AS JSON
,[Time Series (Daily)] NVARCHAR(MAX) AS JSON ) A
CROSS APPLY OPENJSON(A.[Time Series (Daily)]) B
CROSS APPLY OPENJSON(B.[value]) WITH([1. open] DECIMAL(10,4)
,[2. high] DECIMAL(10,4)
--and so on
) C;
简而言之：
我们使用OPENJSON() 和WITH 子句来获得第一级（Meta 和Time Series）。
我们使用AS JSON 告诉引擎，结果仍然是 JSON，而不是简单的文本。
现在我们可以使用JSON_VALUE() 直接读取元数据（这也可以直接针对@JSON）。
我们可以再次使用OPENJSON() 来深入了解。
使用描述性部分作为内容是一个坏习惯。实际上，作为日期值的键最好都是同名并在对象中携带日期......不过我们可以使用OPENJSON 不带 WITH-clause 来检索 key（= 日期）和 value（= 对象）列。
另一个OPENJSON() 获取value 作为输入并将返回内部值。
【讨论】：
太棒了。这是一个巨大的帮助Shnugo以上是关于路径包含有趣字符时的 TSQL OPENJSON的主要内容，如果未能解决你的问题，请参考以下文章