Oracle与DB2、MySQL取前10笔记录的对比-1

 
Oralce的示例：

1. 最佳选择：行使说明函数

row_number() over ( partition by col1 order by col2 )
比方想掏出100-150笔记录,根据tname排序
select tname,tabtype from (
select tname,tabtype,row_number() over ( order by tname ) rn from tab
)
where rn between 100 and 150;
 
2. 运用rownum 虚列

select tname,tabtype from (
select tname,tabtype,rownum rn from tab where rownum <= 150
)
where rn >= 100;

注释：运用序列时不克不及基于整个记录纠集来举办排序,假定指定了order by子句,排序的的是选出来的记录集的排序。
 
create table mynumber(id int,name varchar(10));
insert into mynumber values(1,'no1');
insert into mynumber values(2,'no2');
insert into mynumber values(3,'no3');
insert into mynumber values(4,'no4');
insert into mynumber values(5,'no5');
insert into mynumber values(5,'no6');
insert into mynumber values(6,'no7');
insert into mynumber values(7,'no8');
insert into mynumber values(8,'no9');
insert into mynumber values(9,'no10');
insert into mynumber values(9,'no11');
insert into mynumber values(9,'no12');
insert into mynumber values(10,'no13');
insert into mynumber values(10,'no14');
insert into mynumber values(10,'no15');
insert into mynumber values(11,'no16');
insert into mynumber values(12,'no17');
insert into mynumber values(13,'no18');
 
select id,name,RANK() over ( order by id ) case1,
DENSE_RANK() over ( order by id ) case2,
row_number() over ( order by id ) case3 from mynumber;

(1) 取前10条分例如id记录，假定最后1笔记录的ID依旧有相反的，那么掏出来。

select id,name from mynumber where id in
(select id from (select distinct id
from mynumber) tt where rownum<=10);
select * from (select id,name,RANK()
over ( order by id ) case1,DENSE_RANK()
over ( order by id ) case2,row_number()
over ( order by id ) case3 from mynumber)
tt where case1<=10;
select * from (select id,name,RANK() over
( order by id ) case1,DENSE_RANK() over
( order by id ) case2,row_number() over
( order by id ) case3 from mynumber)
tt where case1 between 5 and 10;
 

(2)取前10笔记录，假定第10笔记录的ID 另有相反的，那么掏出来。

select * from mynumber where id in
(select id from mynumber where rownum <=10);
select * from (select id,name,RANK()
over ( order by id ) case1,DENSE_RANK()
over ( order by id ) case2,row_number()
over ( order by id ) case3 from mynumber)
tt where case2<=10;
select * from (select id,name,RANK()
over ( order by id ) case1,DENSE_RANK()
over ( order by id ) case2,row_number()
over ( order by id ) case3 from mynumber)
tt where case2 between 5 and 10;
 
来自: 新客网(www.xker.com) 详文参考：http://www.xker.com/page/e2008/0128/46627.html


版权声明： 原创作品，许诺转载，转载时请务必以超链接形式标明文章 原始来由 、作者信息和本声明。不然将追究司法责任。