Question

我有SQL脚本，当执行时会在所有数据库上运行，获取所有具有记录计数的表的列表以及在该表上创建的索引。

然而，当我尝试在并行数据仓库服务器上运行它时，它给了我错误。因为并行数据仓库不允许您创建临时表变量，变量等。

我找到了表变量的替代方法，但是没有找到关于声明varchar变量的任何内容，然后主要关注的是脚本使用数据库作为变量的方式，以便在所有数据库上运行脚本。我没有＆＃39;找到任何替代品。

declare @TableList TABLE(Id int IDENTITY(1,1),DataBaseName VARCHAR(100),TableName VARCHAR(100),RecordCount INT,NameOfIndex VARCHAR(100),TypeOfIndex VARCHAR(100))
declare @TableListWithIndex TABLE(Id int IDENTITY(1,1),DataBaseName VARCHAR(100),TableName VARCHAR(100),NameOfIndex VARCHAR(100),TypeOfIndex VARCHAR(100))
declare @DatabaseNames TABLE(Id INT IDENTITY(1,1),NameOfDataBase VARCHAR(100),DataBaseSize VARCHAR(100),Remarks VARCHAR(500))
declare @sql varchar(1000)
INSERT INTO @DatabaseNames 
EXEC sp_databases



DECLARE @Count INT = 0,@Counter INT = 1 ,@NameOfDb VARCHAR(100);

SELECT @Count = COUNT(1) FROM @DatabaseNames 


WHILE(@Count > 0)
BEGIN
    SELECT @NameOfDb = NameOfDataBase FROM @DatabaseNames WHERE Id = @Counter

    SET @sql = N'SELECT '''+ @NameOfDb +''',T.name AS [TABLE NAME], 
           I.rows AS [ROWCOUNT] 
    FROM   @NameOfDb.sys.tables AS T 
           INNER JOIN @NameOfDb.sys.sysindexes AS I 
                   ON T.object_id = I.id 
                      AND I.indid < 2 
    ORDER  BY I.rows DESC';

    SET @sql    =   REPLACE(@sql, '@NameOfDb', @NameOfDb)
    INSERT INTO @TableList(DataBaseName,TableName,RecordCount)
    exec(@sql);




    SET @sql = N'SELECT 
         '''+ @NameOfDb +''',
         TableName = t.name,
         IndexName = ind.name,
         ind.type_desc
    FROM 
         @NameOfDb.sys.indexes ind 
    INNER JOIN 
         @NameOfDb.sys.index_columns ic ON  ind.object_id = ic.object_id and ind.index_id = ic.index_id 
    INNER JOIN 
         @NameOfDb.sys.columns col ON ic.object_id = col.object_id and ic.column_id = col.column_id 
    INNER JOIN 
         @NameOfDb.sys.tables t ON ind.object_id = t.object_id 
    WHERE 
         ind.is_primary_key = 0 
         AND ind.is_unique = 0 
         AND ind.is_unique_constraint = 0 
         AND t.is_ms_shipped = 0 
    ORDER BY 
         t.name, ind.name, ind.index_id, ic.index_column_id ';

    SET @sql    =   REPLACE(@sql, '@NameOfDb', @NameOfDb)
    INSERT INTO @TableListWithIndex(DataBaseName,TableName,NameOfIndex,TypeOfIndex)
    exec(@sql);




    SET @Count = @Count - 1;
    SET @Counter = @Counter + 1;
END

    update TL
     SET TL.NameOfIndex = TLW.NameOfIndex,TL.TypeOfIndex = TLW.TypeOfIndex
     from @TableList TL
     INNER JOIN @TableListWithIndex TLW ON TL.TableName = TLW.TableName 
     AND TL.DataBaseName = TLW.DataBaseName

     select * from @TableList 
     Where DataBaseName NOT IN ('tempdb','ReportServer$MSSQLSERVER2014TempDB','ReportServer$MSSQLSERVER2014','msdb','master','model') 
     order by DataBaseName

现在我如何在并行数据仓库中完成这项工作

Answer 1

您可以使用未记录的sp_MSforeachdb过程在每个数据库上执行脚本。

转换SQL脚本以处理并行数据仓库

1 个答案: