数据库常用SQL:从表中查询N条数据+判断表名/字段名是否存在

从表中查询N条数据:

利用一条SQL语句从数据库Table表中随机获取N条记录，各数据库的SQL语句略有不同，如下：

1、MySql

Select * From TABLE Order By Rand() Limit N

以上代码效率不高，自己对1000条数据表随机取10条的测试结果为耗时0.125s，有网友推荐改为以下代码效率有大幅提高，耗时为0.005s，不过结果是随机从某个位置开始取N条记录，而不是随机取N条记录，略有差异。

SELECT * FROM `TABLE` a join (SELECT ROUND(RAND() * ((SELECT MAX(id) FROM `TABLE`)-(SELECTMIN(id) FROM `TABLE`))+(SELECT MIN(id) FROM `TABLE`)) AS id) AS b WHERE a.id >= b.id ORDER BYa.id LIMIT N;

2、SQL Server

NewID()函数将创建一个 uniqueidentifier 类型的唯一值。

3、Access

Select TOP N * From TABLE Order By Rnd(ID)

Rnd(ID) 其中的ID是自动编号字段，可以利用其他任何数值来完成，比如用姓名字段(UserName)

Select TOP N * From TABLE Order BY Rnd(Len(UserName))

4、Oracle

1）dbms_random包

select * from (select * from Table order by dbms_random.value) where rownum < N;

注：dbms_random包需要手工安装，位于$ORACLE_HOME/rdbms/admin/dbmsrand.sql

dbms_random.value(100,200)

可以产生100到200范围的随机数

2）按概率抽取：

例：

取表Table中记录条数的10%条记录

3）sys_guid

select * from (select * from Table orderby sys_guid()) where rownum<N;

5、postgreSQL

select * from Table order by random() limit N

判断表名/字段名是否存在:

1.MSSQL Server

表:select COUNT(*) from dbo.sysobjectsWHEREname = 'table_name';

字段:select COUNT(*) fromsyscolumnsWHEREid = object_id(‘table_name’) and name= 'column_name';

2.My SQL

表:select COUNT(*) from information_schema.tables WHERE table_name = 'table_name';

字段:select COUNT(*) from information_schema.columns WHERE table_name = 'table_name' and column_name ='column_name';

3.Oracle

表:SELECT count(*) FROM USER_OBJECTS WHERE OBJECT_NAME = 'table_name';

字段:SELECT COUNT(*) FROM USER_TAB_COLUMNS WHERE TABLE_NAME ='table_name'andcolumn_name ='column_name';

4.PostgreSql

表:注意,表名需为小写的

1.selectcount(*) from information_schema.tables WHERE table_schema = 'table_schema' and table_name ='table_name';

2.select count(*) from pg_class where relname = 'qwbddict';

3.select count(*) from pg_statio_user_tables where relname='qwbddict';

字段:select count(*) from information_schema.columns WHERE table_schema = 'table_schema' and table_name ='table_name' and column_name = 'column_name';