Oralce的示例：

1. 最佳选择：利用分析函数

       row_number() over ( partition by col1 order by col2 )
                               比如想取出100-150条记录,按照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. 使用rownum 虚列

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

注释：使用虚列时不能基于整个记录集合来进行排序,假如指定了order by子句,排序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条不同id记录，假如最后1条记录的ID依然有相同的，那么取出来

   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条记录，假如第10条记录的ID 还有相同的，那么取出来

   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;

(3)取前10条记录

       select id,name from mynumber
              where rownum <=10;
               select id,name from
              (select id,name,rownum rn from
                mynumber where rownum <= 10 ) where rn >= 5;

              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 case3<=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 case3 between 5 and 10;