本篇内容主要讲解“Oracle与PostgreSQL子查询有什么不同”，感兴趣的朋友不妨来看看。本文介绍的方法操作简单快捷，实用性强。下面就让小编来带大家学习“Oracle与PostgreSQL子查询有什么不同”吧!
准确的表达应该是在子查询的having条件中出现agg函数且依赖父查询的相关字段时，Oracle支持而PG不支持。
Oracle
创建表，插入数据，执行查询，OK！
TEST-orcl@DESKTOP-V430TU3>drop table tbl1;
Table dropped.
TEST-orcl@DESKTOP-V430TU3>drop table tbl2;
Table dropped.
TEST-orcl@DESKTOP-V430TU3>drop table tbl3;
Table dropped.
TEST-orcl@DESKTOP-V430TU3>
TEST-orcl@DESKTOP-V430TU3>create table tbl1 (id int,c1 int,c2 int,c3 int);
Table created.
TEST-orcl@DESKTOP-V430TU3>create table tbl2 (id int,c1 int,c2 int,c3 int);
Table created.
TEST-orcl@DESKTOP-V430TU3>create table tbl3 (id int,c1 int,c2 int,c3 int);
Table created.
TEST-orcl@DESKTOP-V430TU3>
TEST-orcl@DESKTOP-V430TU3>insert into tbl1 values(1,1,1,1);
1 row created.
TEST-orcl@DESKTOP-V430TU3>insert into tbl1 values(1,1,1,1);
1 row created.
TEST-orcl@DESKTOP-V430TU3>insert into tbl1 values(1,1,1,1);
1 row created.
TEST-orcl@DESKTOP-V430TU3>insert into tbl1 values(2,1,1,1);
1 row created.
TEST-orcl@DESKTOP-V430TU3>insert into tbl1 values(2,1,1,1);
1 row created.
TEST-orcl@DESKTOP-V430TU3>insert into tbl1 values(3,1,1,1);
1 row created.
TEST-orcl@DESKTOP-V430TU3>
TEST-orcl@DESKTOP-V430TU3>insert into tbl2 values(1,1,1,1);
1 row created.
TEST-orcl@DESKTOP-V430TU3>insert into tbl2 values(1,1,1,1);
1 row created.
TEST-orcl@DESKTOP-V430TU3>insert into tbl2 values(1,1,1,1);
1 row created.
TEST-orcl@DESKTOP-V430TU3>insert into tbl2 values(2,1,1,1);
1 row created.
TEST-orcl@DESKTOP-V430TU3>insert into tbl2 values(2,1,1,1);
1 row created.
TEST-orcl@DESKTOP-V430TU3>insert into tbl2 values(3,1,1,1);
1 row created.
TEST-orcl@DESKTOP-V430TU3>
TEST-orcl@DESKTOP-V430TU3>truncate table tbl3;
Table truncated.
TEST-orcl@DESKTOP-V430TU3>insert into tbl3 values(1,1,1,1);
1 row created.
TEST-orcl@DESKTOP-V430TU3>insert into tbl3 values(1,1,1,1);
1 row created.
TEST-orcl@DESKTOP-V430TU3>
TEST-orcl@DESKTOP-V430TU3>commit;
Commit complete.
TEST-orcl@DESKTOP-V430TU3>
TEST-orcl@DESKTOP-V430TU3>truncate table tbl3;
Table truncated.
TEST-orcl@DESKTOP-V430TU3>insert into tbl3 values(1,2,1,1);
1 row created.
TEST-orcl@DESKTOP-V430TU3>
TEST-orcl@DESKTOP-V430TU3>select a.id,sum(a.c1) as sum_c1,sum(a.c2) as sum_c2
2 from tbl1 a,tbl2 b
3 where a.id = b.id
4 and exists (select 1 from tbl3 c where c.id = a.id group by c.id having sum(c.c1) > sum(a.c1))
5 group by a.id;
ID SUM_C1 SUM_C2
---------- ---------- ----------
1 9 9
TEST-orcl@DESKTOP-V430TU3不过，就算Oracle支持这样的写法，也不建议这样来写，原因是SQL语义理解起来并不友好，难以理解。
PG
创建表，插入数据，执行查询，出错。
[pg12@localhost ~]$ psql
Expanded display is used automatically.
psql (12.1)
Type "help" for help.
[local:/data/run/pg12]:5120 pg12@testdb=# drop table tbl1;
s(1,1,1,1);
insert into tbl1 values(1,1,1,1);
insert into tbl2 select * from tbl1;
insert into tbl2 select * from tbl1;
insert into tbl3 select * from tbl1;
commit;
ERROR: table "tbl1" does not exist
[local:/data/run/pg12]:5120 pg12@testdb=# drop table tbl2;
ERROR: table "tbl2" does not exist
[local:/data/run/pg12]:5120 pg12@testdb=# drop table tbl3;
ERROR: table "tbl3" does not exist
[local:/data/run/pg12]:5120 pg12@testdb=#
[local:/data/run/pg12]:5120 pg12@testdb=# create table tbl1 (id int,c1 int,c2 int,c3 int);
CREATE TABLE
[local:/data/run/pg12]:5120 pg12@testdb=# create table tbl2 (id int,c1 int,c2 int,c3 int);
CREATE TABLE
[local:/data/run/pg12]:5120 pg12@testdb=# create table tbl3 (id int,c1 int,c2 int,c3 int);
CREATE TABLE
[local:/data/run/pg12]:5120 pg12@testdb=#
[local:/data/run/pg12]:5120 pg12@testdb=# insert into tbl1 values(1,1,1,1);
INSERT 0 1
[local:/data/run/pg12]:5120 pg12@testdb=# insert into tbl1 values(1,1,1,1);
INSERT 0 1
[local:/data/run/pg12]:5120 pg12@testdb=# insert into tbl1 values(1,1,1,1);
INSERT 0 1
[local:/data/run/pg12]:5120 pg12@testdb=#
[local:/data/run/pg12]:5120 pg12@testdb=# insert into tbl2 select * from tbl1;
INSERT 0 3
[local:/data/run/pg12]:5120 pg12@testdb=# insert into tbl2 select * from tbl1;
INSERT 0 3
[local:/data/run/pg12]:5120 pg12@testdb=#
[local:/data/run/pg12]:5120 pg12@testdb=# insert into tbl3 select * from tbl1;
INSERT 0 3
[local:/data/run/pg12]:5120 pg12@testdb=#
[local:/data/run/pg12]:5120 pg12@testdb=# commit;
WARNING: there is no transaction in progress
COMMIT
[local:/data/run/pg12]:5120 pg12@testdb=# select a.id,sum(a.c1) as sum_c1,sum(a.c2) as sum_c2
pg12@testdb-# from tbl1 a,tbl2 b
pg12@testdb-# where a.id = b.id
pg12@testdb-# and exists (select 1 from tbl3 c where c.id = a.id group by c.id having sum(c.c1) = sum(a.c1))
pg12@testdb-# group by a.id;
ERROR: aggregate functions are not allowed in WHERE
LINE 4: ...ere c.id = a.id group by c.id having sum(c.c1) = sum(a.c1))
^
[local:/data/run/pg12]:5120 pg12@testdb=#出现的错误是“aggregate functions are not allowed in WHERE”，但条件明明在having怎么报WHERE中出现agg函数呢？原因是PG认为条件sum(c.c1) = sum(a.c1)中的a.c1出现在父查询中，该条件认为是WHERE中的条件。
到此，相信大家对“Oracle与PostgreSQL子查询有什么不同”有了更深的了解，不妨来实际操作一番吧！这里是亿速云网站，更多相关内容可以进入相关频道进行查询，关注我们，继续学习！
