表结构如下所示：
亲
CREATE TABLE [ql].[Quantlist] (
[QuantlistId] INT IDENTITY (1,1) NOT NULL,[StateId] INT NOT NULL,[Title] VARCHAR (500) NOT NULL,CONSTRAINT [PK_Quantlist] PRIMARY KEY CLUSTERED ([QuantlistId] ASC),CONSTRAINT [FK_Quantlist_State] FOREIGN KEY ([StateId]) REFERENCES [ql].[State] ([StateId])
);
儿童
CREATE TABLE [ql].[QuantlistAttribute]
(
[QuantlistAttributeId] INT IDENTITY (1,1),[QuantlistId] INT NOT NULL,[Narrative] VARCHAR (500) NOT NULL,CONSTRAINT [PK_QuantlistAttribute] PRIMARY KEY ([QuantlistAttributeId]),CONSTRAINT [FK_QuantlistAttribute_QuantlistId] FOREIGN KEY ([QuantlistId]) REFERENCES [ql].[Quantlist]([QuantlistId]),)
孙
CREATE TABLE [ql].[AttributeReference]
(
[AttributeReferenceId] INT IDENTITY (1,[QuantlistAttributeId] INT NOT NULL,[Reference] VARCHAR (250) NOT NULL,CONSTRAINT [PK_QuantlistReference] PRIMARY KEY ([AttributeReferenceId]),CONSTRAINT [FK_QuantlistReference_QuantlistAttribute] FOREIGN KEY ([QuantlistAttributeId]) REFERENCES [ql].[QuantlistAttribute]([QuantlistAttributeId]),)
在我的存储过程中,我传入QuantlistId,我想要克隆为@QuantlistId.由于QuantlistAttribute表中有一个ForeignKey,所以我也可以很容易克隆.
INSERT INTO [ql].[Quantlist] (
[StateId],[Title],) SELECT
1,Title,FROM [ql].[Quantlist]
WHERE QuantlistId = @QuantlistId
SET @ClonedId = SCOPE_IDENTITY()
INSERT INTO ql.QuantlistAttribute(
QuantlistId,Narrative)
SELECT
@ClonedId,Narrative,FROM ql.QuantlistAttribute
WHERE QuantlistId = @QuantlistId
故障归结于AttributeReference.如果我克隆了30个QuantlistAttribute记录,那么如何克隆参考表中的记录,并将其与刚插入到QuantlistAttribute表中的新记录进行匹配？
INSERT INTO ql.AttributeReference(
QuantlistAttributeId,Reference,)
SELECT
QuantlistAttributeId,FROM ql.QuantlistReference
WHERE ??? I don't have a key to go off of for this.
我以为我可以使用一些临时链接表来保存旧的属性id以及新的属性id.我不知道如何将旧的Attribute Id插入临时表以及它们的新属性.通过QuantlistId插入现有的属性很简单,但是我无法确定如何确保以某种方式将正确的新旧Id连接在一起,以便可以克隆AttributeReference表.如果我可以获得新的和旧的Id链接的QuantlistAttribute,我可以加入该临时表,并找出如何恢复新克隆的引用与新克隆属性的关系.
对此的任何帮助将是非常棒的.我花了最后一天半的时间想出来,没有运气：/
请原谅一些sql不一致.我重新写了sql的真正快速,修剪了很多其他列,相关表和约束,这是不需要这个问题.
编辑
在做一点挖掘之后,我发现OUTPUT可能对此有用.有没有办法使用OUTPUT将我刚插入的QuantlistAttributeId记录映射到他们起源的QuantlistAttributeId？
解决方法
>您可以根据ORDER BY c.QuantlistAttributeId ASC的顺序将数据插入到QuantlistAttribute中
>有一个临时表/表变量3列
> id标识列
>新的QuantlistAttributeId
>旧的QuantlistAttributeId.
>使用OUTPUT将QuantlistAttribute的新标识值插入到临时表/表变量中.
新的ID按照与c.QuantlistAttributeId相同的顺序生成
>使用由QuantlistAttributeId排序的row_number()根据表变量的row_number()和id匹配旧的QuantlistAttributeId和新的QuantlistAttributeIds,并更新表变量中的值或旧的QuantlistAttributeId
>使用临时表并加入AttributeReference并一次性插入记录.
注意：
INSERT INTO SELECT和ROW_NUMBER()中的ORDER BY需要匹配的旧的QuantlistAttributeId,因为查看您的问题,似乎没有其他逻辑键将旧的和新的记录映射到一起.
查询上述步骤
DECLARE @ClonedId INT,@QuantlistId INT = 0
INSERT INTO [ql].[Quantlist] (
[StateId],[Title]
) SELECT
1,Title
FROM [ql].[Quantlist]
WHERE QuantlistId = @QuantlistId
SET @ClonedId = SCOPE_IDENTITY()
--Define a table variable to store the new QuantlistAttributeID and use it to map with the Old QuantlistAttributeID
DECLARE @temp TABLE(id int identity(1,newAttrID INT,oldAttrID INT)
INSERT INTO ql.QuantlistAttribute(
QuantlistId,Narrative)
--New QuantlistAttributeId are created in the same order as old QuantlistAttributeId because of ORDER BY
OUTPUT inserted.QuantlistAttributeId,NULL INTO @temp
SELECT
@ClonedId,Narrative
FROM ql.QuantlistAttribute c
WHERE QuantlistId = @QuantlistId
--This is required to keep new ids generated in the same order as old
ORDER BY c.QuantlistAttributeId ASC
;WITH CTE AS
(
SELECT c.QuantlistAttributeId,--Use ROW_NUMBER to get matching id which is same as the one generated in @temp
ROW_NUMBER()OVER(ORDER BY c.QuantlistAttributeId ASC) id
FROM ql.QuantlistAttribute c
WHERE QuantlistId = @QuantlistId
)
--Update the old value in @temp
UPDATE T
SET oldAttrID = CTE.QuantlistAttributeId
FROM @temp T
INNER JOIN CTE ON T.id = CTE.id
INSERT INTO ql.AttributeReference(
QuantlistAttributeId,Reference)
SELECT
T.NewAttrID,Reference
FROM ql.AttributeReference R
--Use OldAttrID to join with ql.AttributeReference and insert NewAttrID
INNER JOIN @temp T
ON T.oldAttrID = R.QuantlistAttributeId
希望这可以帮助.
