Use

从information_schema中选择*

列出所有的数据库，基本表，sps，视图等在SQL server。

一个不太为人所知的TSQL技术，以随机顺序返回行:

-- Return rows in a random order
SELECT 
    SomeColumn 
FROM 
    SomeTable
ORDER BY 
    CHECKSUM(NEWID())

行构造函数

您可以用一个插入语句插入多行数据。

INSERT INTO Colors (id, Color)
VALUES (1, 'Red'),
       (2, 'Blue'),
       (3, 'Green'),
       (4, 'Yellow')

如果您想删除数据库中的所有过程-

SELECT  IDENTITY ( int, 1, 1 ) id, 
        [name] 
INTO    #tmp 
FROM    sys.procedures 
WHERE   [type]        = 'P' 
    AND is_ms_shipped = 0 

DECLARE @i INT 

SELECT   @i = COUNT( id ) FROM #tmp 
WHILE    @i > 0 
BEGIN 
   DECLARE @name VARCHAR( 100 ) 
   SELECT @name = name FROM #tmp WHERE id = @i 
   EXEC ( 'DROP PROCEDURE ' + @name ) 
   SET @i = @i-1 
END

DROP TABLE #tmp

连接字符串额外功能:

MultipleActiveResultSets = true;

这就产生了ADO。Net 2.0及以上版本在单个数据库连接上读取多个只向前、只读的结果集，如果要进行大量读取，这可以提高性能。即使是混合执行查询类型，也可以打开它。

应用程序名称= MyProgramName

现在，当您希望通过查询sysprocesses表来查看活动连接列表时，程序的名称将出现在program_name列中，而不是“”。Net SqlClient数据提供程序

好吧，这是我的观点:

http://dbalink.wordpress.com/2008/10/24/querying-the-object-catalog-and-information-schema-views/

我懒得在这里重新写一遍，所以请查看我的帖子。这对许多人来说可能是微不足道的，但会有一些人会发现它是一颗“隐藏的宝石”。

编辑:

过了一会儿，我决定在这里添加代码，这样您就不必跳转到我的博客来查看代码了。

SELECT  T.NAME AS [TABLE NAME], C.NAME AS [COLUMN NAME], P.NAME AS [DATA TYPE], P.MAX_LENGTH AS[SIZE],   CAST(P.PRECISION AS VARCHAR) +‘/’+ CAST(P.SCALE AS VARCHAR) AS [PRECISION/SCALE]
FROM ADVENTUREWORKS.SYS.OBJECTS AS T
JOIN ADVENTUREWORKS.SYS.COLUMNS AS C
ON T.OBJECT_ID=C.OBJECT_ID
JOIN ADVENTUREWORKS.SYS.TYPES AS P
ON C.SYSTEM_TYPE_ID=P.SYSTEM_TYPE_ID
WHERE T.TYPE_DESC=‘USER_TABLE’;

或者，如果你想拉出所有的用户表，像这样使用CURSOR:

DECLARE @tablename VARCHAR(60)

DECLARE cursor_tablenames CURSOR FOR
SELECT name FROM AdventureWorks.sys.tables

OPEN cursor_tablenames
FETCH NEXT FROM cursor_tablenames INTO @tablename

WHILE @@FETCH_STATUS = 0
BEGIN

SELECT  t.name AS [TABLE Name], c.name AS [COLUMN Name], p.name AS [DATA Type], p.max_length AS[SIZE],   CAST(p.PRECISION AS VARCHAR) +‘/’+ CAST(p.scale AS VARCHAR) AS [PRECISION/Scale]
FROM AdventureWorks.sys.objects AS t
JOIN AdventureWorks.sys.columns AS c
ON t.OBJECT_ID=c.OBJECT_ID
JOIN AdventureWorks.sys.types AS p
ON c.system_type_id=p.system_type_id
WHERE t.name = @tablename
AND t.type_desc=‘USER_TABLE’
ORDER BY t.name ASC

FETCH NEXT FROM cursor_tablenames INTO @tablename
END

CLOSE cursor_tablenames
DEALLOCATE cursor_tablenames

其他参考资料(我的博客):http://dbalink.wordpress.com/2009/01/21/how-to-create-cursor-in-tsql/