有时候,多么的希望ORACLE能够导出某个视图中的数据,然后通过这个视图来迁移需要的数据,现在ORACLE 12C通过expdp的views_as_tables来实现了该功能,把视图当作一个普通表从而导出数据,导入的时候直接和一个正常表一样,通过视图的导出,表的导入来实现相关需求
准备测试环境

SQL> SELECT * FROM V$VERSION;
BANNER                                                                               CON_ID
-------------------------------------------------------------------------------- ----------
Oracle Database 12c Enterprise Edition Release 12.1.0.0.2 - 64bit                         0
PL/SQL Release 12.1.0.0.2                                                                 0
CORE    12.1.0.0.2                                                                        0
TNS for Linux: Version 12.1.0.0.2                                                         0
NLSRTL Version 12.1.0.0.2
SQL> create table t_lx as select * from v$log;
Table created.
SQL> create table t_xl as select * from v$logfile;
Table created.
SQL> create view v_xifenfei  as
  2   SELECT thread#,
  3           a.sequence#,
  4           a.group#,
  5           TO_CHAR (first_change#, '9999999999999999') "SCN",
  6           a.status,
  7           MEMBER
  8      FROM t_lx a, t_xl b
  9     WHERE a.group# = B.GROUP#
 10  ORDER BY a.sequence# DESC;
View created.
SQL> col member for a50
SQL> set lines 134
SQL> select * from v_xifenfei;
   THREAD#  SEQUENCE#     GROUP# SCN               STATUS           MEMBER
---------- ---------- ---------- ----------------- ---------------- -------------------------------------------
         1         30          3            391892 CURRENT          /u01/app/oracle/oradata/xifenfei/redo03.log
         1         29          2            377363 INACTIVE         /u01/app/oracle/oradata/xifenfei/redo02.log
         1         28          1            374892 INACTIVE         /u01/app/oracle/oradata/xifenfei/redo01.log

expdp结合VIEWS_AS_TABLES导出视图

[oracle@xifenfei ~]$ expdp xff/xifenfei views_as_tables=v_xifenfei directory=data_pump_dir dumpfile=xifenfei.dmp
Export: Release 12.1.0.0.2        on Sun Dec 16 07:56:48 2012
Copyright (c) 1982, 2012, Oracle and/or its affiliates.  All rights reserved.
Connected to: Oracle Database 12c Enterprise Edition Release 12.1.0.0.2 - 64bit
With the Partitioning, OLAP, Data Mining and Real Application Testing options
Starting "XFF"."SYS_EXPORT_TABLE_01":xff/******** views_as_tables=v_xifenfei directory=data_pump_dir
 dumpfile=xifenfei.dmp
Estimate in progress using BLOCKS method...
Processing object type TABLE_EXPORT/VIEWS_AS_TABLES/TABLE_DATA
Total estimation using BLOCKS method: 16 KB
Processing object type TABLE_EXPORT/VIEWS_AS_TABLES/TABLE
. . exported "XFF"."V_XIFENFEI"                          7.390 KB       3 rows
Master table "XFF"."SYS_EXPORT_TABLE_01" successfully loaded/unloaded
******************************************************************************
Dump file set for XFF.SYS_EXPORT_TABLE_01 is:
  /u01/app/oracle/admin/xifenfei/dpdump/xifenfei.dmp
Job "XFF"."SYS_EXPORT_TABLE_01" successfully completed at Sun Dec 16 07:58:17 2012 elapsed 0 00:00:56

impdp导入数据

[oracle@xifenfei ~]$ impdp xff/xifenfei remap_table=v_xifenfei:v_xff directory=data_pump_dir dumpfile=xifenfei.dmp
Import: Release 12.1.0.0.2        on Sun Dec 16 08:06:06 2012
Copyright (c) 1982, 2012, Oracle and/or its affiliates.  All rights reserved.
Connected to: Oracle Database 12c Enterprise Edition Release 12.1.0.0.2 - 64bit
With the Partitioning, OLAP, Data Mining and Real Application Testing options
Master table "XFF"."SYS_IMPORT_FULL_01" successfully loaded/unloaded
Starting "XFF"."SYS_IMPORT_FULL_01":xff/******** remap_table=v_xifenfei:v_xff directory=data_pump_dir
 dumpfile=xifenfei.dmp
Processing object type TABLE_EXPORT/VIEWS_AS_TABLES/TABLE
Processing object type TABLE_EXPORT/VIEWS_AS_TABLES/TABLE_DATA
. . imported "XFF"."V_XFF"                               7.390 KB       3 rows
Job "XFF"."SYS_IMPORT_FULL_01" successfully completed at Sun Dec 16 08:06:20 2012 elapsed 0 00:00:10

验证数据

SQL> COL OBJECT_NAME FOR A20
SQL> select object_type,object_name from user_objectS where object_name like 'V_X%';
OBJECT_TYPE             OBJECT_NAME
----------------------- --------------------
VIEW                    V_XIFENFEI
TABLE                   V_XFF
SQL> col member for a50
SQL> set lines 134
SQL> select * from v_XFF;
   THREAD#  SEQUENCE#     GROUP# SCN               STATUS           MEMBER
---------- ---------- ---------- ----------------- ---------------- --------------------------------------------
         1         30          3            391892 CURRENT          /u01/app/oracle/oradata/xifenfei/redo03.log
         1         29          2            377363 INACTIVE         /u01/app/oracle/oradata/xifenfei/redo02.log
         1         28          1            374892 INACTIVE         /u01/app/oracle/oradata/xifenfei/redo01.log

通过测试证明在12C中ORACLE的expdp/impdp可以实现导出视图数据,进入导入到[其他库]其他表中

对11gR2比较熟悉的朋友应该比较清楚,在该版本中引入了一个新的特性,能够快速的增加一个新列,具体见Oracle 11g增加列,并带默认值的新特性,但是这个功能在该版本中总有个不足,需要设置默认值,在有些情况下,有些列就是不需要默认值,在12C的版本中,解决了这个鸡肋,能够快速增加一个列而且可以是不指定默认值(默认值为NULL)
数据库12C版本

SQL> select * from v$version;
BANNER                                                                               CON_ID
-------------------------------------------------------------------------------- ----------
Oracle Database 12c Enterprise Edition Release 12.1.0.0.2 - 64bit                         0
PL/SQL Release 12.1.0.0.2                                                                 0
CORE    12.1.0.0.2                                                                        0
TNS for Linux: Version 12.1.0.0.2                                                         0
NLSRTL Version 12.1.0.0.2                                                                 0

创建模拟表

SQL> create table t_xifenfei(id number,name varchar2(20));
Table created.
SQL> desc t_xifenfei
 Name                                      Null?    Type
 ----------------------------------------- -------- ----------------------------
 ID                                                 NUMBER
 NAME                                               VARCHAR2(20)
SQL> insert into t_xifenfei values (1,'www.xifenfei.com');
1 row created.
SQL> insert into t_xifenfei values (2,'www.xifenfei.com');
1 row created.
SQL> insert into t_xifenfei values (3,'www.xifenfei.com');
1 row created.
SQL> commit;
Commit complete.
SQL> select * from t_xifenfei;
        ID NAME
---------- --------------------
         1 www.xifenfei.com
         2 www.xifenfei.com
         3 www.xifenfei.com

第一次dump block

SQL> select   rowid,
  2   dbms_rowid.rowid_relative_fno(rowid)rel_fno,
  3   dbms_rowid.rowid_block_number(rowid)blockno,
  4   dbms_rowid.rowid_row_number(rowid) rowno
  5   from t_xifenfei ;
ROWID                 REL_FNO    BLOCKNO      ROWNO
------------------ ---------- ---------- ----------
AAAEy3AAEAAAAGGAAA          4        390          0
AAAEy3AAEAAAAGGAAB          4        390          1
AAAEy3AAEAAAAGGAAC          4        390          2
SQL> alter system dump datafile 4 block 390;
System altered.
--dump block
block_row_dump:
tab 0, row 0, @0x1f81
tl: 23 fb: --H-FL-- lb: 0x1  cc: 2
col  0: [ 2]  c1 02
col  1: [16]  77 77 77 2e 78 69 66 65 6e 66 65 69 2e 63 6f 6d
tab 0, row 1, @0x1f6a
tl: 23 fb: --H-FL-- lb: 0x1  cc: 2
col  0: [ 2]  c1 03
col  1: [16]  77 77 77 2e 78 69 66 65 6e 66 65 69 2e 63 6f 6d
tab 0, row 2, @0x1f53
tl: 23 fb: --H-FL-- lb: 0x1  cc: 2
col  0: [ 2]  c1 04
col  1: [16]  77 77 77 2e 78 69 66 65 6e 66 65 69 2e 63 6f 6d
end_of_block_dump

增加不含默认值列

SQL> alter table t_xifenfei add c_xff varchar2(100);
Table altered.
SQL> desc t_xifenfei
 Name                                      Null?    Type
 ----------------------------------------- -------- ----------------------------
 ID                                                 NUMBER
 NAME                                               VARCHAR2(20)
 C_XFF                                              VARCHAR2(100)
--证明增加列无默认值
SQL> insert into t_xifenfei values(4,'www.xifenfei.com','www.orasos.com');
1 row created.
SQL> commit;
Commit complete.

第二次dump block

SQL> alter system checkpoint;
System altered.
SQL>  alter system dump datafile 4 block 390;
System altered.
--block dump
block_row_dump:
tab 0, row 0, @0x1f81
tl: 23 fb: --H-FL-- lb: 0x1  cc: 2
col  0: [ 2]  c1 02
col  1: [16]  77 77 77 2e 78 69 66 65 6e 66 65 69 2e 63 6f 6d
tab 0, row 1, @0x1f6a
tl: 23 fb: --H-FL-- lb: 0x1  cc: 2
col  0: [ 2]  c1 03
col  1: [16]  77 77 77 2e 78 69 66 65 6e 66 65 69 2e 63 6f 6d
tab 0, row 2, @0x1f53
tl: 23 fb: --H-FL-- lb: 0x1  cc: 2
col  0: [ 2]  c1 04
col  1: [16]  77 77 77 2e 78 69 66 65 6e 66 65 69 2e 63 6f 6d
tab 0, row 3, @0x1f2d
tl: 38 fb: --H-FL-- lb: 0x2  cc: 3
col  0: [ 2]  c1 05
col  1: [16]  77 77 77 2e 78 69 66 65 6e 66 65 69 2e 63 6f 6d
col  2: [14]  77 77 77 2e 6f 72 61 73 6f 73 2e 63 6f 6d
end_of_block_dump

从这里可以明显的看出来,前面的三条记录只有2列,但是四条记录有3列,证明使用了11gR2的新特性,这里可以使用null的默认值,证明比以往版本新特性增强.

在ORACLE 12C中有了Invisible Columns的概念,就是在表中真实的存在该列,但是通过设置Invisible导致该列不可显示.官方说明:

The property of whether a column is visible can be controlled by the user.
Invisible columns are not seen unless specified explicitly in the SELECT list.
Any generic access of a table (such as a SELECT * FROM table or a DESCRIBE)
  will not show invisible columns.

数据库版本

SQL> select * from v$version;
BANNER                                                                               CON_ID
-------------------------------------------------------------------------------- ----------
Oracle Database 12c Enterprise Edition Release 12.1.0.0.2 - 64bit                         0
PL/SQL Release 12.1.0.0.2                                                                 0
CORE    12.1.0.0.2                                                                        0
TNS for Linux: Version 12.1.0.0.2                                                         0
NLSRTL Version 12.1.0.0.2                                                                 0

创建含INVISIBLE列表

SQL> CREATE TABLE t_xifenfei (a number, b number INVISIBLE, c number);
Table created.
SQL> desc t_xifenfei
 Name                                                  Null?    Type
 ----------------------------------------------------- -------- ---------------------------
 A                                                              NUMBER
 C                                                              NUMBER
SQL>  select column_name,HIDDEN_COLUMN,COLUMN_ID from user_TAB_COLs  where TABLE_NAME='T_XIFENFEI';
COLUMN_NAM HID  COLUMN_ID
---------- --- ----------
A          NO           1
B          YES
C          NO           2

通过观察可以发现INVISIBLE列在一般的查询中不显示,在USER_TAB_COLS的视图中显示:HIDDEN_COLUMN为YES而且COLUMN_ID为空

设置INVISIBLE列为VISIBLE

SQL> ALTER TABLE t_xifenfei MODIFY (b VISIBLE);
Table altered.
SQL> desc t_xifenfei
 Name                                      Null?    Type
 ----------------------------------------- -------- ----------------------------
 A                                                  NUMBER
 C                                                  NUMBER
 B                                                  NUMBER
SQL> select column_name,HIDDEN_COLUMN,COLUMN_ID from user_TAB_COLs  where TABLE_NAME='T_XIFENFEI';
COLUMN_NAM HID  COLUMN_ID
---------- --- ----------
A          NO           1
B          NO           3
C          NO           2

当设置为VISIBLE时HIDDEN_COLUMN为YES而且COLUMN_ID为递增值

修改列展示顺序(Column Ordering)

SQL> ALTER TABLE t_xifenfei MODIFY (a invisible);
Table altered.
SQL> ALTER TABLE t_xifenfei MODIFY (a visible);
Table altered.
SQL> desc t_xifenfei
 Name                                      Null?    Type
 ----------------------------------------- -------- ----------------------------
 C                                                  NUMBER
 B                                                  NUMBER
 A                                                  NUMBER
SQL> select column_name,HIDDEN_COLUMN,COLUMN_ID from user_TAB_COLs  where TABLE_NAME='T_XIFENFEI';
COLUMN_NAM HID  COLUMN_ID
---------- --- ----------
A          NO           3
B          NO           2
C          NO           1

通过INVISIBLE和VISIBLE相关操作,实现A列从头移尾,实现列的Column Ordering效果.
补充说明:在以前的blog中,提供了修改col$基表的方法(通过修改col$.col#改变列展示顺序)来实现列的顺序修改,相对于这种方法来说,修改数据字典的方法风险太大,需要非常谨慎,而且不被ORACLE SUPPORT

ORACLE 12C在分区维护方面有了不少的增强,在12C的beta版本中已经支持多分区的add/truncate/drop/merge操作,大大的提高了分区维护的效率.
数据库版本

SQL> select * from v$version;
BANNER                                                                               CON_ID
-------------------------------------------------------------------------------- ----------
Oracle Database 12c Enterprise Edition Release 12.1.0.0.2 - 64bit                         0
PL/SQL Release 12.1.0.0.2                                                                 0
CORE    12.1.0.0.2                                                                        0
TNS for Linux: Version 12.1.0.0.2                                                         0
NLSRTL Version 12.1.0.0.2                                                                 0

在FF PDB中创建xff用户

SQL> alter session set container=ff;
Session altered.
SQL> create user xff identified by xifenfei;
User created.
SQL> grant dba to xff;
Grant succeeded.
SQL> conn xff/xifenfei@ff
Connected.
SQL> show con_name;
CON_NAME
------------------------------
FF

创建分区表RANGE PARTITIONS

SQL> CREATE TABLE t_xifenfei
  2  (name varchar2(100),time_id DATE)
  3  partition by range(time_id)
  4  (partition xff_2006 values less than (TO_DATE('01-01-2007','dd-MM-yyyy')),
  5  partition xff_2007 values less than (TO_DATE('01-01-2008','dd-MM-yyyy')),
  6  partition xff_2008 values less than (TO_DATE('01-01-2009','dd-MM-yyyy')),
  7  partition xff_2009 values less than (TO_DATE('01-01-2010','dd-MM-yyyy')));
Table created.
SQL> SET LONG 30
SQL> select PARTITION_NAME,HIGH_VALUE FROM USER_TAB_PARTITIONS where table_name='T_XIFENFEI';
PARTITION_NAME                 HIGH_VALUE
------------------------------ ------------------------------
XFF_2006                       TO_DATE(' 2007-01-01 00:00:00'
XFF_2007                       TO_DATE(' 2008-01-01 00:00:00'
XFF_2008                       TO_DATE(' 2009-01-01 00:00:00'
XFF_2009                       TO_DATE(' 2010-01-01 00:00:00'

ADD 多个分区

SQL> ALTER TABLE t_xifenfei ADD
  2    PARTITION XFF_2010 VALUES LESS THAN (TO_DATE('01-01-2011','dd-MM-yyyy')),
  3    PARTITION XFF_2011 VALUES LESS THAN (TO_DATE('01-01-2012','dd-MM-yyyy')),
  4    PARTITION XFF_2012 VALUES LESS THAN (TO_DATE('01-01-2013','dd-MM-yyyy'));
Table altered.
SQL> select PARTITION_NAME,HIGH_VALUE FROM USER_TAB_PARTITIONS where table_name='T_XIFENFEI';
PARTITION_NAME                 HIGH_VALUE
------------------------------ ------------------------------
XFF_2006                       TO_DATE(' 2007-01-01 00:00:00'
XFF_2007                       TO_DATE(' 2008-01-01 00:00:00'
XFF_2008                       TO_DATE(' 2009-01-01 00:00:00'
XFF_2009                       TO_DATE(' 2010-01-01 00:00:00'
XFF_2010                       TO_DATE(' 2011-01-01 00:00:00'
XFF_2011                       TO_DATE(' 2012-01-01 00:00:00'
XFF_2012                       TO_DATE(' 2013-01-01 00:00:00'
7 rows selected.

Split多个分区

SQL> ALTER TABLE t_xifenfei  split PARTITION  XFF_2012 INTO
  2    (PARTITION XFF_2012_03 VALUES LESS THAN (TO_DATE('01-03-2012','dd-MM-yyyy')),
  3    PARTITION XFF_2012_06 VALUES LESS THAN (TO_DATE('01-06-2012','dd-MM-yyyy')),
  4    PARTITION XFF_2012_09 VALUES LESS THAN (TO_DATE('01-09-2012','dd-MM-yyyy')),
  5    PARTITION XFF_2012);
Table altered.
SQL>  select PARTITION_NAME,HIGH_VALUE FROM USER_TAB_PARTITIONS where table_name='T_XIFENFEI';
PARTITION_NAME                 HIGH_VALUE
------------------------------ ------------------------------
XFF_2006                       TO_DATE(' 2007-01-01 00:00:00'
XFF_2007                       TO_DATE(' 2008-01-01 00:00:00'
XFF_2008                       TO_DATE(' 2009-01-01 00:00:00'
XFF_2009                       TO_DATE(' 2010-01-01 00:00:00'
XFF_2010                       TO_DATE(' 2011-01-01 00:00:00'
XFF_2011                       TO_DATE(' 2012-01-01 00:00:00'
XFF_2012                       TO_DATE(' 2013-01-01 00:00:00'
XFF_2012_03                    TO_DATE(' 2012-03-01 00:00:00'
XFF_2012_06                    TO_DATE(' 2012-06-01 00:00:00'
XFF_2012_09                    TO_DATE(' 2012-09-01 00:00:00'
10 rows selected.

插入分区数据

SQL>INSERT INTO t_xifenfei VALUES('www.xifenfei.com',SYSDATE-100);
1 row created.
SQL> INSERT INTO t_xifenfei VALUES('www.xifenfei.com',SYSDATE-200);
1 row created.
SQL> INSERT INTO t_xifenfei VALUES('www.xifenfei.com',SYSDATE-300);
1 row created.
SQL>  INSERT INTO t_xifenfei VALUES('www.xifenfei.com',SYSDATE-10);
1 row created.
SQL> commit;
Commit complete.
SQL> col name for a20
SQL> select * from t_xifenfei;
NAME                 TIME_ID
-------------------- ---------
www.xifenfei.com     17-FEB-12
www.xifenfei.com     27-MAY-12
www.xifenfei.com     04-SEP-12
www.xifenfei.com     03-DEC-12
SQL>select * from t_xifenfei PARTITION(XFF_2012_03);
NAME                 TIME_ID
-------------------- ---------
www.xifenfei.com     17-FEB-12
SQL> select * from t_xifenfei PARTITION(XFF_2012_06);
NAME                 TIME_ID
-------------------- ---------
www.xifenfei.com     27-MAY-12
SQL>  select * from t_xifenfei PARTITION(XFF_2012_09);
no rows selected
SQL>  select * from t_xifenfei PARTITION(XFF_2012);
NAME                 TIME_ID
-------------------- ---------
www.xifenfei.com     04-SEP-12
www.xifenfei.com     03-DEC-12

TRUNCATE 多个分区

SQL> Alter table t_xifenfei truncate partitions XFF_2012_03, XFF_2012_06, XFF_2012_09;
Table truncated.
--剩下两条记录存在于XFF_2012中
SQL> select * from t_xifenfei;
NAME                 TIME_ID
-------------------- ---------
www.xifenfei.com     04-SEP-12
www.xifenfei.com     03-DEC-12
SQL> SELECT SUBOBJECT_NAME,object_id,data_object_id from user_objects where SUBOBJECT_NAME like 'XFF_2012_0%';
SUBOBJECT_NAME                  OBJECT_ID DATA_OBJECT_ID
------------------------------ ---------- --------------
XFF_2012_09                         90603          90603    <---为什么没有变
XFF_2012_06                         90602          90606
XFF_2012_03                         90601          90605
SQL> select PARTITION_NAME,HIGH_VALUE,SEGMENT_CREATED FROM USER_TAB_PARTITIONS where table_name='T_XIFENFEI';
PARTITION_NAME                 HIGH_VALUE                     SEGM
------------------------------ ------------------------------ ----
XFF_2006                       TO_DATE(' 2007-01-01 00:00:00' NO
XFF_2007                       TO_DATE(' 2008-01-01 00:00:00' NO
XFF_2008                       TO_DATE(' 2009-01-01 00:00:00' NO
XFF_2009                       TO_DATE(' 2010-01-01 00:00:00' NO
XFF_2010                       TO_DATE(' 2011-01-01 00:00:00' NO
XFF_2011                       TO_DATE(' 2012-01-01 00:00:00' NO
XFF_2012                       TO_DATE(' 2013-01-01 00:00:00' YES
XFF_2012_03                    TO_DATE(' 2012-03-01 00:00:00' YES
XFF_2012_06                    TO_DATE(' 2012-06-01 00:00:00' YES
XFF_2012_09                    TO_DATE(' 2012-09-01 00:00:00' NO
--XFF_2012_09因为块延迟创建,没有segment导致truncate对应的dataobj#不变
10 rows selected.

DROP 多个分区

SQL>  Alter table t_xifenfei DROP  partitions XFF_2012_03, XFF_2012_06, XFF_2012_09;
Table altered.
SQL> select PARTITION_NAME,HIGH_VALUE FROM USER_TAB_PARTITIONS where table_name='T_XIFENFEI';
PARTITION_NAME                 HIGH_VALUE
------------------------------ ------------------------------
XFF_2006                       TO_DATE(' 2007-01-01 00:00:00'
XFF_2007                       TO_DATE(' 2008-01-01 00:00:00'
XFF_2008                       TO_DATE(' 2009-01-01 00:00:00'
XFF_2009                       TO_DATE(' 2010-01-01 00:00:00'
XFF_2010                       TO_DATE(' 2011-01-01 00:00:00'
XFF_2011                       TO_DATE(' 2012-01-01 00:00:00'
XFF_2012                       TO_DATE(' 2013-01-01 00:00:00'
7 rows selected.

MERGE 多分区

SQL> Alter table t_xifenfei merge partitions XFF_2006, XFF_2007, XFF_2008 into partition XFF_OLD;
Table altered.
SQL> select PARTITION_NAME,HIGH_VALUE FROM USER_TAB_PARTITIONS where table_name='T_XIFENFEI';
PARTITION_NAME                 HIGH_VALUE
------------------------------ ------------------------------
XFF_2009                       TO_DATE(' 2010-01-01 00:00:00'
XFF_2010                       TO_DATE(' 2011-01-01 00:00:00'
XFF_2011                       TO_DATE(' 2012-01-01 00:00:00'
XFF_2012                       TO_DATE(' 2013-01-01 00:00:00'
XFF_OLD                        TO_DATE(' 2009-01-01 00:00:00'

本测试是基于Range partitions进行,其实在ORACLE 12C中对于分区表的维护做了比较大的增强,上面试验的多分区操作,也支持List partitions和subpartitions.ddl一次性操作多个分区,给分区经常做维护的DBA来说,带来了不少的方便,省去了很多重复行工作.