[code=SQL]create table bloodtype
(
  id          VARCHAR2(40) not null,
  orgcode     VARCHAR2(200) not null,
  orgname     VARCHAR2(200) not null,
  ABO         VARCHAR2(40) not null,
  amount      VARCHAR2(20),
  creatdate   VARCHAR2(400)
)

create table bloodtype
(
  id       VARCHAR2(40) not null,
  name     VARCHAR2(200) not null,
  description VARCHAR2(400)
)

--這個意思?你的第二個表bloodtype寫錯了吧,我用bloodtypedtl代替的,值就用A,B,AB,O代替的
select 
    bloodtype.orgname,    
    sum(decode(bloodtypedtl.name,'A',bloodtype.amount,0)) "A型血数量",
    sum(decode(bloodtypedtl.name,'B',bloodtype.amount,0)) "B型血数量",
    sum(decode(bloodtypedtl.name,'AB',bloodtype.amount,0)) "AB型血数量",
    sum(decode(bloodtypedtl.name,'O',bloodtype.amount,0)) "O型血数量"
from bloodtype,bloodtypedtl
where bloodtype.ABO=bloodtypedtl.id
group by bloodtype.orgname;

------解决方案--------------------


with t as (
select '001' as code, 'A' as abo,12 as amount from dual
union all
select '001','B',22 from dual
union all
select '001','AB',32 from dual
union all
select '001','O',21 from dual
union all
select '002','B',45 from dual
union all
select '002','AB',54 from dual
union all
select '002','O',44 from dual
union all
select '002','A',55 from dual
)
select * from t
pivot (sum(amount) for abo in ('A','B','AB','O'));
CODE 'A'                    'B'                    'AB'                   'O'                   
---- ---------------------- ---------------------- ---------------------- ----------------------  
002  55                     45                     54                     44                    
001  12                     22                     32                     21