oracle pivot，unpivot基本语法

适用版本，oracle11及以上。

pivot

老方法

用于行转列。在11g之前，一般用这样的方法来做

对于这样的表

select 1 id, 'a1' typ, 1 val from dual union all
select 1, 'a2' , 2 val from dual union all
select 1, 'a3' , 3 val from dual union all
select 2, 'a1' , 4 val from dual union all
select 2, 'a3' , 5 val from dual 

ID	TYP	VAL
1	a1	1
1	a2	2
1	a3	3
2	a1	4
2	a3	5

我现在想要将每一个id所对应的typ横着显示，需要这样写

with tab1 as (
select 1 id, 'a1' typ, 1 val from dual union all
select 1, 'a2' , 2 val from dual union all
select 1, 'a3' , 3 val from dual union all
select 2, 'a1' , 4 val from dual union all
select 2, 'a3' , 5 val from dual 
)
select id,
       max(decode(typ, 'a1', val)) a1,
       max(decode(typ, 'a2', val)) a2,
       max(decode(typ, 'a3', val)) a3
  from tab1
 group by id
;

ID	A1	A2	A3
1	1	2	3
2	4		5

新方法

适用pivot，可以这样写

with tab1 as (
select 1 id, 'a1' typ, 1 val from dual union all
select 1, 'a2' , 2 val from dual union all
select 1, 'a3' , 3 val from dual union all
select 2, 'a1' , 4 val from dual union all
select 2, 'a3' , 5 val from dual 
)
select*from tab1
pivot(max(val) for typ in('a1' as aaa, 'a2' as a2, 'a3'))
;

ID	AAA	A2	‘a3’
1	1	2	3
2	4		5

unpivot

重功能上来说，可以看成是pivot的逆运算，只是行列变换所需的列会由于手动指定的关系而缺失或者改变一些值。

with tab1 as (
select 1 id, 'a1' typ, 1 val from dual union all
select 1, 'a2' , 2 val from dual union all
select 1, 'a3' , 3 val from dual union all
select 2, 'a1' , 4 val from dual union all
select 2, 'a3' , 5 val from dual 
)
select*from tab1
pivot(max(val) for typ in('a1' as aaa, 'a2' as a2, 'a3' a3))
unpivot(val for typ in (aaa as 'a', a2 as 'b', a3 as 'c'))
;

ID	TYP	VAL
1	a	1
1	b	2
1	c	3
2	a	4
2	c	5

xml

经常看到有人问怎么把in中的部分写成动态的，我通过官方文档得知，对于以xml格式存储的数据是可以做到的。不过从来没遇到的那样的表，先不研究了。