关于零售业中的二八sql统计

零售业里的，20%商品，产生80%的销售，就是要查出是哪些商品的贡献度最大。
如图：
我如何取每类商品中按销售额降序排列后，依次累加直到累计值>=0.8*sum（销售额）的前几种商品的销售额明细。

sql：
-- Create table
create table TAB_1
(
  COD VARCHAR2(10),
  REC NUMBER(8,4)
);
insert into TAB_1 (COD, REC)
values ('A', 500);
insert into TAB_1 (COD, REC)
values ('B', 600);
insert into TAB_1 (COD, REC)
values ('C', 200);
insert into TAB_1 (COD, REC)
values ('D', 223);
insert into TAB_1 (COD, REC)
values ('D', 30);
insert into TAB_1 (COD, REC)
values ('A', 55);
insert into TAB_1 (COD, REC)
values ('A', 3);
commit;

elect m.R1, m.Cod, m.Rec1, n.Rec1 需要查询的数量, n.Cod 需要查询的代码
  From (Select Cod, Rec1, Rownum R1
          From (Select Cod, Rec1
                  From (Select Cod, Sum(Rec) Rec1 From Tab_1 a Group By Cod)
                 Order By Rec1)) m,
       (Select Cod, Rec1, Rownum R2
          From (Select Cod, Rec1
                  From (Select Cod, Sum(Rec) Rec1 From Tab_1 a Group By Cod)
                 Order By Rec1)) n
Where m.R1 <= n.R2
   And m.R1 =
       (Select Max(R1)
          From (Select m.R1, Sum(n.Rec1)
                  From (Select Cod, Rec1, Rownum R1
                          From (Select Cod, Rec1
                                  From (Select Cod, Sum(Rec) Rec1
                                          From Tab_1 a
                                         Group By Cod)
                                 Order By Rec1)) m,
                       (Select Cod, Rec1, Rownum R2
                          From (Select Cod, Rec1
                                  From (Select Cod, Sum(Rec) Rec1
                                          From Tab_1 a
                                         Group By Cod)
                                 Order By Rec1)) n
                 Where m.R1 <= n.R2
                 Group By m.R1
                Having Sum(n.Rec1) > = (Select Sum(Rec) * 0.8 From Tab_1)))
Order By m.Rec1 Desc

SQL也不是万事万灵的，
LZ提供的这个场景，用单纯的SQL来实现，
只能得到一个理解困难，效率低下，难于维护的东东
可以考虑把一部分SQL不好实现的逻辑写到程序中

用以下两个SQL文作检索

--查询1
SELECT 
	  X.REC_SUM
	, X.Cod
FROM 
(
	SELECT 
		  SUM(Rec) AS REC_SUM
		, Cod
	FROM Tab_1 a 
	Group By Cod
) X
ORDER BY X.REC_SUM DESC
/

--查询2
SELECT SUM(Rec) AS ALL_REC_SUM
FROM Tab_1
/

然后在程序中对查询1的结果进行迭代，对REC_SUM进行累计，
当累计值大于等于ALL_REC_SUM的80%的时候，迭代停止

同意楼上，谢谢关注；

偶只是变态的想通过一个语句来实现，看看到底能实现不

嘿嘿 这是我的t-sql实现。
同意这个功能在程序中实现更好。 
写这个sql只1为了练习sql。 希望能有更多的人参与， 大家交流一下嘛。

select rownm=identity(12), COD, sum(REC) total 
  into #recsum 
  from TAB_1 
  group by COD 
  order by total des

select COD 
  from #recsum
  where rownm <= (
   select min(t.rownm) from 
       (select rownm, COD, total, (select sum(total) from #recsum where rownm <= o.rownm) totalUntilNow 
          from #recsum o) t 
       where t.totalUntilNow > (select sum(REC)*0.8 from TAB_1))

输出：
COD
----------
B
A
D