Table 具有主键更改的值参数

Table valued parameter with primary keys changes

我刚开始在 sql 服务器 2008 中使用 table 值参数来更新 table。基本上,我逐列创建用户定义的 table 类型映射到现有 table,并将其用作 table 值参数,将数据传递到服务器以进行删除、更新和插入。在大多数情况下,效果很好。但是,对于具有复合主键的 table,当部分键发生更改时,此方法将失败,因为 tvp 仅包含一组键值(当前或原始键值)。我的问题是处理此类案件的最佳方式是什么?我是否必须在 table 类型定义中添加第二组主键列?

一个简化的例子: 表格:

CREATE TABLE [dbo].[Prices](
    [ID] [int] NOT NULL PRIMARY KEY,
    [Date] [smalldatetime] NOT NULL PRIMARY KEY,
    [Value] [float] NOT NULL,
CONSTRAINT [PK_Prices] PRIMARY KEY CLUSTERED 
(
    [ID] ASC,
    [Date] ASC
)
)

Table Type
CREATE TYPE [dbo].[tvp_Prices] AS TABLE(
    [ID] [int] NOT NULL,
    [Date] [smalldatetime] NOT NULL,
    [Value] [float] NOT NULL
)

Add some data to the table
Insert Into  [dbo].[Prices] Values (1, '1/1/2015', 1.2)
Insert Into  [dbo].[Prices] Values (1, '1/2/2015', 1.3)
Insert Into  [dbo].[Prices] Values (2, '1/1/2015', 1.4)
Insert Into  [dbo].[Prices] Values (2, '1/2/2015', 1.5)

in c#, load the Prices to DataTable pricesTable,

update some value:
pricesTable.Rows[1]["Date"] = new DateTime(2015, 1, 3);

现在的问题是如何使用 table 值参数 (tvp_Prices) 将更改保存回服务器?

Table值参数用于将多行数据通过参数传递给例如存储过程。在您的情况下,用例可能是下一个:

CREATE TYPE [dbo].[tvp_Prices] AS TABLE(
    [ID] [int] NOT NULL,
    [Date] [smalldatetime] NOT NULL,
    [Value] [float] NOT NULL
);

CREATE TABLE [dbo].[Prices](
    [ID] [int] NOT NULL PRIMARY KEY,
    [Date] [smalldatetime] NOT NULL PRIMARY KEY,
    [Value] [float] NOT NULL,
    CONSTRAINT [PK_Prices] PRIMARY KEY CLUSTERED 
    (
        [ID] ASC,
        [Date] ASC
    )
);

CREATE PROC dbo.use_Prices
(
    @prices [dbo].[tvp_Prices] READONLY
)
AS BEGIN
    INSERT INTO [dbo].[Prices] ([ID], [Date], [Value])
    SELECT * FROM @prices;
END


-- usage ---------------------------------

DECLARE
    @prices [dbo].[tvp_Prices];

Insert Into  @prices Values (1, '1/1/2015', 1.2)
Insert Into  @prices Values (1, '1/2/2015', 1.3)
Insert Into  @prices Values (2, '1/1/2015', 1.4)
Insert Into  @prices Values (2, '1/2/2015', 1.5)

SELECT * FROM [dbo].[Prices]

EXEC dbo.use_Prices @prices;

SELECT * FROM [dbo].[Prices]

结果

ID  Date    Value
-----------------------------

-

ID  Date                Value
-----------------------------
1   2015-01-01 00:00:00 1.2
1   2015-01-02 00:00:00 1.3
2   2015-01-01 00:00:00 1.4
2   2015-01-02 00:00:00 1.5

更新示例

CREATE PROC dbo.use_UpdatePrices
(
    @prices [dbo].[tvp_Prices] READONLY
)
AS BEGIN
    -- update values by the PK
    UPDATE p
    SET
        p.[Value] = p1.[Value]
    FROM [dbo].[Prices] p
    JOIN @prices p1 ON p1.ID = p.[ID]
    AND p1.[Date] = p.[Date];
END

Truncate table [dbo].[Prices]

Insert Into  [dbo].[Prices] Values (1, '1/1/2015', 1.2)
Insert Into  [dbo].[Prices] Values (1, '1/2/2015', 1.3)
Insert Into  [dbo].[Prices] Values (2, '1/1/2015', 1.4)
Insert Into  [dbo].[Prices] Values (2, '1/2/2015', 1.5)

-- usage -----------------------------------
DECLARE
@prices [dbo].[tvp_Prices];

-- values to update
Insert Into  @prices Values (1, '1/1/2015', 10.10) 
Insert Into  @prices Values (1, '1/2/2015', 11.11)

SELECT * FROM [dbo].[Prices]

EXEC dbo.use_UpdatePrices @prices;

SELECT * FROM [dbo].[Prices]

输出

ID  Date                Value
-----------------------------
1   2015-01-01 00:00:00 1.2
1   2015-01-02 00:00:00 1.3
2   2015-01-01 00:00:00 1.4
2   2015-01-02 00:00:00 1.5


ID  Date                Value
-----------------------------
1   2015-01-01 00:00:00 10.1
1   2015-01-02 00:00:00 11.11
2   2015-01-01 00:00:00 1.4
2   2015-01-02 00:00:00 1.5