semijoin链接进行subquery unnesting.
drop table emp1;
drop table dept1;
create table emp1 as select * from emp;
create table dept1 as select * from dept;
SQL> select e.empno, e.deptno
from emp1 e
where e.deptno in (select d.deptno from dept1 d where d.loc = 'CHICAGO'); 2 3
EMPNO DEPTNO
---------- ----------
7900 30
7844 30
7698 30
7654 30
7521 30
7499 30
6 rows selected.
SQL> select e.empno, e.deptno
from emp1 e, dept1 d
where e.deptno = d.deptno
and d.loc = 'CHICAGO'; 2 3 4
EMPNO DEPTNO
---------- ----------
7499 30
7521 30
7654 30
7698 30
7844 30
7900 30
6 rows selected.
此时子查询被改写成关联,结果完全等价,是因为d.deptno上是Unique索引
那如果d.deptno不唯一呢？
SQL> select * from dept1;
DEPTNO DNAME LOC
---------- -------------- -------------
10 ACCOUNTING NEW YORK
20 RESEARCH DALLAS
30 SALES CHICAGO
40 OPERATIONS BOSTON
30 SALES CHICAGO
SQL> select e.empno, e.deptno
from emp1 e
where e.deptno in (select d.deptno from dept1 d where d.loc = 'CHICAGO'); 2 3
EMPNO DEPTNO
---------- ----------
7900 30
7844 30
7698 30
7654 30
7521 30
7499 30
6 rows selected.
SQL> select e.empno, e.deptno
from emp1 e, dept1 d
where e.deptno = d.deptno
and d.loc = 'CHICAGO'; 2 3 4
EMPNO DEPTNO
---------- ----------
7499 30
7499 30
7521 30
7521 30
7654 30
7654 30
7698 30
7698 30
7844 30
7844 30
7900 30
EMPNO DEPTNO
---------- ----------
7900 30
12 rows selected.
改写成关联后结果就翻倍了得去从
SQL> select e.empno, e.deptno
from emp1 e, dept1 d
where e.deptno = d.deptno
and d.loc = 'CHICAGO'
group by e.empno, e.deptno 2 3 4 5 ;
EMPNO DEPTNO
---------- ----------
7844 30
7521 30
7698 30
7900 30
7654 30
7499 30
6 rows selected.