SQLServer ——将存储过程从一个数据库复制到另一个数据库

我是 SQL 的新手,我需要做的是组合2。将 mdf 数据库合并为一个。我使用 SQLServer2008 Manager-Tasks > Import/Export 表完成了这项工作。表和视图已成功复制,但新数据库中没有存储过程。有什么办法吗?

173678 次浏览
  • Right click on database
  • Tasks
  • Generate Scripts
  • Select the objects you wish to script
  • Script to File
  • Run generated scripts against target database

You can use SSMS's "Generate Scripts..." function to script out whatever you need to transfer. Right-click on the source database in SSMS, choose "Generate Scripts...", and follow the wizard along. Then run your resultant script that will now contain the stored procedure create statements.

In Mgmt Studio, right-click on your original database then Tasks then Generate Scripts... - follow the wizard.

use

select * from sys.procedures

to show all your procedures;

sp_helptext @objname = 'Procedure_name'

to get the code

and your creativity to build something to loop through them all and generate the export code :)

You can generate scriptof the stored proc's as depicted in other answers. Once the script have been generated, you can use sqlcmd to execute them against target DB like

sqlcmd -S <server name> -U <user name> -d <DB name> -i <script file> -o <output log file>

Late one but gives more details that might be useful…

Here is a list of things you can do with advantages and disadvantages

Generate scripts using SSMS

  • Pros: extremely easy to use and supported by default
  • Cons: scripts might not be in the correct execution order and you might get errors if stored procedure already exists on secondary database. Make sure you review the script before executing.

Third party tools

  • Pros: tools such as ApexSQL Diff (this is what I use but there are many others like tools from Red Gate or Dev Art) will compare two databases in one click and generate script that you can execute immediately
  • Cons: these are not free (most vendors have a fully functional trial though)

System Views

  • Pros: You can easily see which stored procedures exist on secondary server and only generate those you don’t have.
  • Cons: Requires a bit more SQL knowledge

Here is how to get a list of all procedures in some database that don’t exist in another database

select *
from DB1.sys.procedures P
where P.name not in
(select name from DB2.sys.procedures P2)

SELECT definition + char(13) + 'GO' FROM MyDatabase.sys.sql_modules s INNER JOIN MyDatabase.sys.procedures p ON [s].[object_id] = [p].[object_id] WHERE p.name LIKE 'Something%'" queryout "c:\SP_scripts.sql -S MyInstance -T -t -w

get the sp and execute it

This code copies all stored procedures in the Master database to the target database, you can copy just the procedures you like by filtering the query on procedure name.

@sql is defined as nvarchar(max), @Name is the target database

DECLARE c CURSOR FOR
SELECT Definition
FROM [ResiDazeMaster].[sys].[procedures] p
INNER JOIN [ResiDazeMaster].sys.sql_modules m ON p.object_id = m.object_id


OPEN c


FETCH NEXT FROM c INTO @sql


WHILE @@FETCH_STATUS = 0
BEGIN
SET @sql = REPLACE(@sql,'''','''''')
SET @sql = 'USE [' + @Name + ']; EXEC(''' + @sql + ''')'


EXEC(@sql)


FETCH NEXT FROM c INTO @sql
END


CLOSE c
DEALLOCATE c

I originally found this post looking for a solution to copying stored procedures from my remote production database to my local development database. After success using the suggested approach in this thread, I realized I grew increasingly lazy (or resourceful, whichever you prefer) and wanted this to be automated. I came across this link, which proved to be very helpful (thank you vincpa), and I extended upon it, resulting in the following file (schema_backup.ps1):

$server             = "servername"
$database           = "databaseName"
$output_path        = "D:\prod_schema_backup"
$login = "username"
$password = "password"


$schema             = "dbo"
$table_path         = "$output_path\table\"
$storedProcs_path   = "$output_path\stp\"
$views_path         = "$output_path\view\"
$udfs_path          = "$output_path\udf\"
$textCatalog_path   = "$output_path\fulltextcat\"
$udtts_path         = "$output_path\udtt\"


[System.Reflection.Assembly]::LoadWithPartialName("Microsoft.SqlServer.ConnectionInfo")  | out-null
[System.Reflection.Assembly]::LoadWithPartialName("Microsoft.SqlServer.SMO") | out-null
[System.Reflection.Assembly]::LoadWithPartialName("Microsoft.SqlServer.SmoExtended")  | out-null
$srvConn = new-object Microsoft.SqlServer.Management.Common.ServerConnection
$srvConn.ServerInstance = $server
$srvConn.LoginSecure = $false
$srvConn.Login = $login
$srvConn.Password = $password
$srv        = New-Object Microsoft.SqlServer.Management.SMO.Server($srvConn)
$db         = New-Object ("Microsoft.SqlServer.Management.SMO.Database")
$tbl        = New-Object ("Microsoft.SqlServer.Management.SMO.Table")
$scripter   = New-Object Microsoft.SqlServer.Management.SMO.Scripter($srvConn)


# Get the database and table objects
$db = $srv.Databases[$database]


$tbl            = $db.tables | Where-object { $_.schema -eq $schema  -and -not $_.IsSystemObject }
$storedProcs    = $db.StoredProcedures | Where-object { $_.schema -eq $schema -and -not $_.IsSystemObject }
$views          = $db.Views | Where-object { $_.schema -eq $schema }
$udfs           = $db.UserDefinedFunctions | Where-object { $_.schema -eq $schema -and -not $_.IsSystemObject }
$catlog         = $db.FullTextCatalogs
$udtts          = $db.UserDefinedTableTypes | Where-object { $_.schema -eq $schema }


# Set scripter options to ensure only data is scripted
$scripter.Options.ScriptSchema  = $true;
$scripter.Options.ScriptData    = $false;


#Exclude GOs after every line
$scripter.Options.NoCommandTerminator   = $false;
$scripter.Options.ToFileOnly            = $true
$scripter.Options.AllowSystemObjects    = $false
$scripter.Options.Permissions           = $true
$scripter.Options.DriAllConstraints     = $true
$scripter.Options.SchemaQualify         = $true
$scripter.Options.AnsiFile              = $true


$scripter.Options.SchemaQualifyForeignKeysReferences = $true


$scripter.Options.Indexes               = $true
$scripter.Options.DriIndexes            = $true
$scripter.Options.DriClustered          = $true
$scripter.Options.DriNonClustered       = $true
$scripter.Options.NonClusteredIndexes   = $true
$scripter.Options.ClusteredIndexes      = $true
$scripter.Options.FullTextIndexes       = $true


$scripter.Options.EnforceScriptingOptions   = $true


function CopyObjectsToFiles($objects, $outDir) {
#clear out before
Remove-Item $outDir* -Force -Recurse
if (-not (Test-Path $outDir)) {
[System.IO.Directory]::CreateDirectory($outDir)
}


foreach ($o in $objects) {


if ($o -ne $null) {


$schemaPrefix = ""


if ($o.Schema -ne $null -and $o.Schema -ne "") {
$schemaPrefix = $o.Schema + "."
}


#removed the next line so I can use the filename to drop the stored proc
#on the destination and recreate it
#$scripter.Options.FileName = $outDir + $schemaPrefix + $o.Name + ".sql"
$scripter.Options.FileName = $outDir + $schemaPrefix + $o.Name
Write-Host "Writing " $scripter.Options.FileName
$scripter.EnumScript($o)
}
}
}


# Output the scripts
CopyObjectsToFiles $tbl $table_path
CopyObjectsToFiles $storedProcs $storedProcs_path
CopyObjectsToFiles $views $views_path
CopyObjectsToFiles $catlog $textCatalog_path
CopyObjectsToFiles $udtts $udtts_path
CopyObjectsToFiles $udfs $udfs_path


Write-Host "Finished at" (Get-Date)
$srv.ConnectionContext.Disconnect()

I have a .bat file that calls this, and is called from Task Scheduler. After the call to the Powershell file, I have:

for /f %f in ('dir /b d:\prod_schema_backup\stp\') do sqlcmd /S localhost /d dest_db /Q "DROP PROCEDURE %f"

That line will go thru the directory and drop the procedures it is going to recreate. If this wasn't a development environment, I would not like programmatically dropping procedures this way. I then rename all the stored procedure files to have .sql:

powershell Dir d:\prod_schema_backup\stp\ | Rename-Item -NewName { $_.name + ".sql" }

And then run:

for /f %f in ('dir /b d:\prod_schema_backup\stp\') do sqlcmd /S localhost /d dest_db /E /i "%f".sql

And that iterates through all the .sql files and recreates the stored procedures. I hope that any part of this will prove to be helpful to someone.

Another option is to transfer stored procedures using SQL Server Integration Services (SSIS). There is a task called Transfer SQL Server Objects Task. You can use the task to transfer the following items:

  • Tables
  • Views
  • Stored Procedures
  • User-Defined Functions
  • Defaults
  • User-Defined Data Types
  • Partition Functions
  • Partition Schemes
  • Schemas
  • Assemblies
  • User-Defined Aggregates
  • User-Defined Types
  • XML Schema Collection

It's a graphical tutorial for Transfer SQL Server Objects Task.

-- This program copies (CREATE OR ALTER) a single PROCEDURE from one database to another
declare @SourceDatabase nvarchar(50);
declare @SourceSchemaName nvarchar(50)
declare @TargetDatabase nvarchar(50);
declare @ProceduresName nvarchar(50);
declare @sql nvarchar(max)




 

set @SourceDatabase =   N'Northwind'        -- The name of the source database
set @SourceSchemaName = N'dbo'              -- The name of Procedure  SCHEME
set @ProceduresName =   N'CustOrderHist'    -- The name of Procedure
set @TargetDatabase =   N'AdventureWorks'   -- The name of the Target database
-- -------- - - -
 

-- If the PROCEDURE SCHEME does not exist, create it
set @sql = ' use [' +@TargetDatabase +'] ' +
' IF NOT EXISTS (SELECT * FROM sys.schemas WHERE name = ''' + @SourceSchemaName+''') '+
' BEGIN ' +
'   EXEC('' CREATE SCHEMA '+ @SourceSchemaName +''') ' +
' END'
 

exec (@sql);


set @sql = ''
--
set @sql = @sql + ' use [' + @TargetDatabase +'] ;' +
' declare @sql2 nvarchar(max) ;' +
' SELECT @sql2 = coalesce(@sql2,'';'' ) + [definition] + '' ; '' ' +
' FROM  ['+@sourceDatabase+'].[sys].[procedures] p '  +
' INNER JOIN ['+@sourceDatabase+'].sys.sql_modules m ON p.object_id = m.object_id '+
' where SCHEMA_NAME([schema_id]) = ''' +@SourceSchemaName +''' and [name] = N''' + @ProceduresName + '''  ; ' +
' set @sql2 = replace(@sql2,''CREATE PROCEDURE'',''CREATE OR ALTER PROCEDURE'')' +
' exec (@sql2)'
exec (@sql)