如何为包含所有存储行的现有 SQL Server 表生成 INSERT 脚本？

Posted 2023-02-15

【中文标题】如何为包含所有存储行的现有 SQL Server 表生成 INSERT 脚本？

【英文标题】：How can I generate an INSERT script for an existing SQL Server table that includes all stored rows?

【发布时间】：2011-07-01 05:22:09

【问题描述】：

我正在寻找一种使用 SQL Management Studio 2008 R2 生成“创建并插入所有行”脚本的方法。

我知道我可以创建一个“创建表”脚本。

我也可以创建一个“插入”脚本，但这只会生成带有占位符的单行。

有没有办法生成一个包含所有当前存储行的插入脚本？

【问题讨论】：

What is the best way to auto-generate INSERT statements for a SQL Server table?的可能重复

【参考方案1】：

是的，但您需要在数据库级别运行它。

在 SSMS 中右键单击数据库，选择“任务”、“生成脚本...”。在您完成工作时，您将进入“脚本选项”部分。点击“高级”，然后在弹出的列表中显示“脚本的数据类型”，您可以选择数据和/或架构。

【讨论】：

不错！正如 gbn 建议的那样，我曾经依赖工具包插件来完成此类任务。很高兴知道我不再需要它了。

【参考方案2】：

此脚本生成现有数据的插入语句。这是一个存储过程，您需要运行一次，然后为您量身定制。

我找了一阵子这种东西，但对结果不满意，所以写了这个存储过程。

例子：

Exec [dbo].[INS]  'Dbo.test where 1=1'

(1) 这里dbo 是模式，test 是表名，1=1 是条件。

Exec [dbo].[INS]  'Dbo.test where name =''neeraj''' * for string

(2) 这里dbo是schema，test是tablename，name='neeraj'是condition。

这是存储过程

/*
Authore : neeraj prasad sharma (please dont remove this :))
Example (1) Exec [dbo].[INS]  'Dbo.test where 1=1'
        (2) Exec [dbo].[INS]  'Dbo.test where name =''neeraj''' * for string

here Dbo is schema and test is tablename and 1=1 is condition

*/


CREATE procedure  [dbo].[INS]                              
(                                                          
   @Query  Varchar(MAX)                                                          
)                              

AS                              

SET nocount ON                  

DECLARE @WithStrINdex as INT                            
DECLARE @WhereStrINdex as INT                            
DECLARE @INDExtouse as INT                            

DECLARE @SchemaAndTAble VArchar(270)                            
DECLARE @Schema_name  varchar(30)                            
DECLARE @Table_name  varchar(240)                            
DECLARE @Condition  Varchar(MAX)                             

SET @WithStrINdex=0                            

SELECT @WithStrINdex=CHARINDEX('With',@Query )                            
, @WhereStrINdex=CHARINDEX('WHERE', @Query)                            

IF(@WithStrINdex!=0)                            
SELECT @INDExtouse=@WithStrINdex                            
ELSE                            
SELECT @INDExtouse=@WhereStrINdex                            

SELECT @SchemaAndTAble=Left (@Query,@INDExtouse-1)                                                     
SELECT @SchemaAndTAble=Ltrim (Rtrim( @SchemaAndTAble))                            

SELECT @Schema_name= Left (@SchemaAndTAble, CharIndex('.',@SchemaAndTAble )-1)                            
,      @Table_name = SUBSTRING(  @SchemaAndTAble , CharIndex('.',@SchemaAndTAble )+1,LEN(@SchemaAndTAble) )                            

,      @CONDITION=SUBSTRING(@Query,@WhereStrINdex+6,LEN(@Query))--27+6                            


DECLARE @COLUMNS  table (Row_number SmallINT , Column_Name VArchar(Max) )                              
DECLARE @CONDITIONS as varchar(MAX)                              
DECLARE @Total_Rows as SmallINT                              
DECLARE @Counter as SmallINT              

DECLARE @ComaCol as varchar(max)            
SELECT @ComaCol=''                   

SET @Counter=1                              
SET @CONDITIONS=''                              

INSERT INTO @COLUMNS                              
SELECT Row_number()Over (Order by ORDINAL_POSITION ) [Count], Column_Name 
FROM INformation_schema.columns 
WHERE Table_schema=@Schema_name AND table_name=@Table_name         


SELECT @Total_Rows= Count(1) 
FROM @COLUMNS                              

SELECT @Table_name= '['+@Table_name+']'                      

SELECT @Schema_name='['+@Schema_name+']'                      

While (@Counter<=@Total_Rows )                              
begin                               
--PRINT @Counter                              

SELECT @ComaCol= @ComaCol+'['+Column_Name+'],'            
FROM @COLUMNS                              
WHERE [Row_number]=@Counter                          

SELECT @CONDITIONS=@CONDITIONS+ ' + Case When ['+Column_Name+'] is null then ''Null'' Else '''''''' + Replace( Convert(varchar(Max),['+Column_Name+']  ) ,'''''''',''''  ) +'''''''' end+'+''','''                                                     
FROM @COLUMNS                              
WHERE [Row_number]=@Counter                              

SET @Counter=@Counter+1                              

End                              

SELECT @CONDITIONS=Right(@CONDITIONS,LEN(@CONDITIONS)-2)                              

SELECT @CONDITIONS=LEFT(@CONDITIONS,LEN(@CONDITIONS)-4)              
SELECT @ComaCol= substring (@ComaCol,0,  len(@ComaCol) )                            

SELECT @CONDITIONS= '''INSERT INTO '+@Schema_name+'.'+@Table_name+ '('+@ComaCol+')' +' Values( '+'''' + '+'+@CONDITIONS                              

SELECT @CONDITIONS=@CONDITIONS+'+'+ ''')'''                              

SELECT @CONDITIONS= 'Select  '+@CONDITIONS +'FRom  ' +@Schema_name+'.'+@Table_name+' With(NOLOCK) ' + ' Where '+@Condition                              
print(@CONDITIONS)                              
Exec(@CONDITIONS)  

【讨论】：

这是一个非常有用的脚本。是否有一种简单的方法可以更改日期时间格式，以便以标准格式导出日期，例如 "2014-03-31 23:55:41" ？目前我得到的日期时间类似于“Feb 10 2010 2:44PM”。

看起来你必须去掉 SELECT FROM 部分并去掉方括号才能使 SPROC 工作。

没有计算列检查。如果您的表有任何计算列，则会出现错误

很好的答案，谢谢 Neeraj，对我来说小错误“INformation_schema”应该是“INFORMATION_SCHEMA”和“DECLARE Con​​dition Varchar(MAX)”应该是“DECLARE CONDITION Varchar(MAX)”

【参考方案3】：

是的，使用商业但便宜的SSMS Tools Pack addin，它有一个漂亮的“从结果集、表或数据库生成插入语句”功能

【讨论】：

2012 版不再免费。

【参考方案4】：

只是为了分享，我已经开发了自己的脚本来做到这一点。随意使用它。它会生成“SELECT”语句，然后您可以在表上运行这些语句以生成“INSERT”语句。

select distinct 'SELECT ''INSERT INTO ' + schema_name(ta.schema_id) + '.' + so.name + ' (' + substring(o.list, 1, len(o.list)-1) + ') VALUES ('
+ substring(val.list, 1, len(val.list)-1) + ');''  FROM ' + schema_name(ta.schema_id) + '.' + so.name + ';'
from    sys.objects so
join sys.tables ta on ta.object_id=so.object_id
cross apply
(SELECT '  ' +column_name + ', '
 from information_schema.columns c
 join syscolumns co on co.name=c.COLUMN_NAME and object_name(co.id)=so.name and OBJECT_NAME(co.id)=c.TABLE_NAME and co.id=so.object_id and c.TABLE_SCHEMA=SCHEMA_NAME(so.schema_id)
 where table_name = so.name
 order by ordinal_position
FOR XML PATH('')) o (list)
cross apply
(SELECT '''+' +case
         when data_type = 'uniqueidentifier' THEN 'CASE WHEN [' + column_name+'] IS NULL THEN ''NULL'' ELSE ''''''''+CONVERT(NVARCHAR(MAX),[' + COLUMN_NAME + '])+'''''''' END '
         WHEN data_type = 'timestamp' then '''''''''+CONVERT(NVARCHAR(MAX),CONVERT(BINARY(8),[' + COLUMN_NAME + ']),1)+'''''''''
         WHEN data_type = 'nvarchar' then 'CASE WHEN [' + column_name+'] IS NULL THEN ''NULL'' ELSE ''''''''+REPLACE([' + COLUMN_NAME + '],'''''''','''''''''''')+'''''''' END'
         WHEN data_type = 'varchar' then 'CASE WHEN [' + column_name+'] IS NULL THEN ''NULL'' ELSE ''''''''+REPLACE([' + COLUMN_NAME + '],'''''''','''''''''''')+'''''''' END'
         WHEN data_type = 'char' then 'CASE WHEN [' + column_name+'] IS NULL THEN ''NULL'' ELSE ''''''''+REPLACE([' + COLUMN_NAME + '],'''''''','''''''''''')+'''''''' END'
         WHEN data_type = 'nchar' then 'CASE WHEN [' + column_name+'] IS NULL THEN ''NULL'' ELSE ''''''''+REPLACE([' + COLUMN_NAME + '],'''''''','''''''''''')+'''''''' END'
         when DATA_TYPE='datetime' then 'CASE WHEN [' + column_name+'] IS NULL THEN ''NULL'' ELSE ''''''''+CONVERT(NVARCHAR(MAX),[' + COLUMN_NAME + '],121)+'''''''' END '
         when DATA_TYPE='datetime2' then 'CASE WHEN [' + column_name+'] IS NULL THEN ''NULL'' ELSE ''''''''+CONVERT(NVARCHAR(MAX),[' + COLUMN_NAME + '],121)+'''''''' END '
         when DATA_TYPE='date' then 'CASE WHEN [' + column_name+'] IS NULL THEN ''NULL'' ELSE ''''''''+CONVERT(NVARCHAR(MAX),[' + COLUMN_NAME + '],121)+'''''''' END '
         when DATA_TYPE='datetimeoffset' then 'CASE WHEN [' + column_name+'] IS NULL THEN ''NULL'' ELSE ''''''''+CONVERT(NVARCHAR(MAX),[' + COLUMN_NAME + '],121)+'''''''' END '
         when DATA_TYPE='geography' and column_name<>'Shape' then 'ST_GeomFromText(''POINT('+column_name+'.Lat '+column_name+'.Long)'') '
         when DATA_TYPE='geography' and column_name='Shape' then '''''''''+CONVERT(NVARCHAR(MAX),[' + COLUMN_NAME + '])+'''''''''
         when DATA_TYPE='bit' then '''''''''+CONVERT(NVARCHAR(MAX),[' + COLUMN_NAME + '])+'''''''''
         when DATA_TYPE='xml' then 'CASE WHEN [' + column_name+'] IS NULL THEN ''NULL'' ELSE ''''''''+REPLACE(CONVERT(NVARCHAR(MAX),[' + COLUMN_NAME + ']),'''''''','''''''''''')+'''''''' END '
         WHEN DATA_TYPE='image' then 'CASE WHEN [' + column_name+'] IS NULL THEN ''NULL'' ELSE ''''''''+CONVERT(NVARCHAR(MAX),CONVERT(VARBINARY(MAX),[' + COLUMN_NAME + ']),1)+'''''''' END '
         WHEN DATA_TYPE='varbinary' then 'CASE WHEN [' + column_name+'] IS NULL THEN ''NULL'' ELSE ''''''''+CONVERT(NVARCHAR(MAX),[' + COLUMN_NAME + '],1)+'''''''' END '
         WHEN DATA_TYPE='binary' then 'CASE WHEN [' + column_name+'] IS NULL THEN ''NULL'' ELSE ''''''''+CONVERT(NVARCHAR(MAX),[' + COLUMN_NAME + '],1)+'''''''' END '
         when DATA_TYPE='time' then 'CASE WHEN [' + column_name+'] IS NULL THEN ''NULL'' ELSE ''''''''+CONVERT(NVARCHAR(MAX),[' + COLUMN_NAME + '])+'''''''' END '
         ELSE 'CASE WHEN [' + column_name+'] IS NULL THEN ''NULL'' ELSE CONVERT(NVARCHAR(MAX),['+column_name+']) END' end
   + '+'', '
 from information_schema.columns c
 join syscolumns co on co.name=c.COLUMN_NAME and object_name(co.id)=so.name and OBJECT_NAME(co.id)=c.TABLE_NAME and co.id=so.object_id and c.TABLE_SCHEMA=SCHEMA_NAME(so.schema_id)
 where table_name = so.name
 order by ordinal_position
FOR XML PATH('')) val (list)
where   so.type = 'U'