SQL Server 2016 for JSON输出整数数组

时间:2016-05-28 00:54:51

标签: arrays sql-server json sql-server-2016

我想使用SQL Server 2016的For JSON功能获取带有整数数组的JSON。我在整数数组上难以接受。

数据库表结构:

declare @Employees table (ID int, Name nvarchar(50))
insert into @Employees values
(1, 'Bob'),
(2, 'Randy')

declare @Permissions table (ID int, PermissionName nvarchar(50))
insert into @Permissions values
(1, 'Post'),
(2, 'Comment'),
(3, 'Edit'),
(4, 'Delete')

declare @EmployeePermissions table (EmployeeID int, PermissionID int)
insert into @EmployeePermissions values
(1, 1),
(1, 2),
(2, 1),
(2, 2),
(2, 3)

期望的结果:

{"EmployeePermissions": [
  {"Employee":"Bob", "Permissions":[1,2]},
  {"Employee":"Randy", "Permissions":[1,2,3]}
}

这是我最接近的,但不是我想要的。

select
    e.Name as Employee,
    (select 
         convert(nvarchar(10),ep.PermissionID) as PermID 
     from @EmployeePermissions ep 
     where ep.EmployeeID=e.ID 
     for json path) as 'Permissions'
from
    @Employees e
for json path, root('EmployeePermissions')

返回:

{"EmployeePermissions": [
  {"Employee":"Bob", "Permissions":[{"permID":1},{"permID":2}]},
  {"Employee":"Randy", "Permissions":[{"permID":1},{"permID":2},{"permID":3}]}
}

2 个答案:

答案 0 :(得分:9)

在AdventureWorks 2016 CTP3 JSON示例中,您可以找到一个可以清除键:值对的数组并创建数组od值的函数:

DROP FUNCTION IF EXISTS dbo.ufnToRawJsonArray
GO
CREATE FUNCTION
[dbo].[ufnToRawJsonArray](@json nvarchar(max), @key nvarchar(400)) returns nvarchar(max)
AS BEGIN
       declare @new nvarchar(max) = replace(@json, CONCAT('},{"', @key,'":'),',')
       return '[' + substring(@new, 1 + (LEN(@key)+5), LEN(@new) -2 - (LEN(@key)+5)) + ']'
END

只需将SELECT FOR JSON表达式的结果作为@json参数提供,并将要删除的键的名称作为第二个参数提供。可能类似于:

select
e.Name as Employee,
JSON_QUERY(dbo.ufnToRawJsonArray(
    (select 
     convert(nvarchar(10),ep.PermissionID) as PermID 
     from @EmployeePermissions ep 
     where ep.EmployeeID=e.ID 
     for json path)
  , 'PermID'))
   as 'Permissions'
from
@Employees e
for json path, root('EmployeePermissions')

答案 1 :(得分:2)

您可以使用FOR XML PATHSTUFF为每个PermissionID分隔Employee个字符串逗号,在其上使用QUOTENANE,然后将所有变量放入变量并将"[替换为[,将]"替换为]

DECLARE @json NVARCHAR(max)

SELECT @json = REPLACE(REPLACE((
    SELECT  e.Name as [Employee],
            QUOTENAME(STUFF((SELECT ','+CAST(ep.PermissionID as nvarchar(10))
            FROM EmployeePermissions ep
            WHERE e.ID = ep.EmployeeID
            FOR XML PATH('')),1,1,''))
            as [Permissions]
    FROM Employees e 
    FOR JSON AUTO, ROOT('EmployeePermissions')
),'"[','['),']"',']')

SELECT @json

输出:

{"EmployeePermissions":[
    {"Employee":"Bob","Permissions":[1,2]},
    {"Employee":"Randy","Permissions":[1,2,3]}
]}

修改

另一种方式:

SELECT '{"EmployeePermissions":[' + STUFF((
SELECT  ',{"Employee":"' + e.Name + '","Permissions":[' +
        STUFF((SELECT ',' + CAST(PermissionID as nvarchar(10))
        FROM EmployeePermissions ep
        WHERE ep.EmployeeID = e.ID
        FOR XML PATH('')),1,1,'') +']}'
FROM Employees e
FOR XML PATH('')),1,1,'') + ']}'

输出:

{"EmployeePermissions":[
    {"Employee":"Bob","Permissions":[1,2]},
    {"Employee":"Randy","Permissions":[1,2,3]}
]}