有一个哥们提出一个问题：
有个表，创建时候的列顺序是a,b,c
如何使用select * 的时候，让列的显示顺序是a,c,b 而且任性地必须使用select *来查询，且不能重建表。
假设有个表test
SQL> select * from test; A B C -------------------- -------------------- -------------------- you pig are SQL> select a,c,b from test; A C B -------------------- -------------------- -------------------- you are pig SQL>
按照其要求，应该是以下显示方式：
SQL> select * from test; A C B -------------------- -------------------- -------------------- you are pig SQL>
可以通过修改sys.col$表来改变表中列的顺序
SQL> SELECT owner,object_name,object_id FROM ALL_OBJECTS WHERE OBJECT_NAME=‘TEST‘;
OWNER OBJECT_NAME OBJECT_ID
------------------------------ ------------------------------ ----------
SYS TEST 88842
SQL> select obj#,col#,name from sys.col$ where obj# =88842 order by col#;
OBJ# COL# NAME
---------- ---------- ------------------------------
88842 1 A
88842 2 B
88842 3 C
SQL> update sys.col$ set col#=2 where obj# =88842 and name=‘C‘;
1 row updated.
SQL> update sys.col$ set col#=3 where obj# =88842 and name=‘B‘;
1 row updated.
SQL> commit;
Commit complete.
SQL> select * from test;
A C B
-------------------- -------------------- --------------------
you are pig
SQL>
原文：http://www.cnblogs.com/abclife/p/5608699.html