情景

原来表中的列定义成VARCHAR2类型，众所周知，VARCHAR2类型最大支持长度为4000。假设因为业务须要。想把此列转换为CLOB类型，在Oracle中直接通过ALTER语句转换是行不通的。以下依据详细事例解说在Oracle数据库中怎样把表列由VARCHAR2类型转换为CLOB类型。

演示样例准备

1. 新建两张张表TB_WITHOUT_DATA（此VARCHAR2列不包括数据）和TB_WITH_DATA（此Varchar2列包括数据)

create table TB_WITHOUT_DATA
(
  id NUMBER,
  name VARCHAR2(100),
  description VARCHAR2(2000)
);

create table TB_WITH_DATA
(
  id NUMBER,
  name VARCHAR2(100),
  description VARCHAR2(2000)
);

insert into TB_WITH_DATA VALUES (1,'David Louis','He is capable of resolving such kind of issue');
insert into TB_WITH_DATA VALUES (2,'German Noemi','She is very beatiful and charming');
insert into TB_WITH_DATA VALUES (3,'Oliver Queen','He is main actor in the Green Arrow');
insert into TB_WITH_DATA VALUES (4,'Mark Williams','He plays snooker very well');
insert into TB_WITH_DATA VALUES (5,'Sita Rama Raju Kata','I do not know this guy');
insert into TB_WITH_DATA VALUES (6,'Promethus','This is a very nice movie');
commit;

错误方法

ALTER TABLE TB_WITHOUT_DATA MODIFY description clob;

错误信息：

SQL Error: ORA-22858: invalid alteration of datatype
22858. 00000 -  "invalid alteration of datatype"
*Cause:    An attempt was made to modify the column type to object, REF,
           nested table, VARRAY or LOB type.
*Action:   Create a new column of the desired type and copy the current
           column data to the new type using the appropriate type
           constructor.

解决方法

 方法一：对于此列没有数据的可通过下面方法改动-首先把该列改成Long类型，然后再改成clob类型

alter table TB_WITHOUT_DATA modify description long;--首先改成Long类型
alter table TB_WITHOUT_DATA modify description clob;--在Long类型的基础上改成clob类型

注：对于此列已经存在数据的。不能通过此方法。否则会报例如以下错误：

alter table TB_WITH_DATA modify description long;--更改包括数据的列

SQL Error: ORA-01439: column to be modified must be empty to change datatype
01439. 00000 -  "column to be modified must be empty to change datatype"

方法二：此方法适合此列包括数据和此列不包括数据两种情况

步骤一：把原来表中该列重命名

alter table TB_WITHOUT_DATA rename column description to description_bak;
alter table TB_WITH_DATA rename column description to description_bak;

步骤二：在表中添加该列，并指定改列类型为clob

alter table TB_WITHOUT_DATA add description clob;
alter table TB_WITH_DATA add description clob;

步骤三：对此列包括数据的须要包数据从步骤一重命名列中拷出(对于此列没有数据的此步骤省略)

update TB_WITH_DATA set description=description_bak;
commit;

步骤四：删除步骤一中的备份列

alter table TB_WITHOUT_DATA drop column description_bak;
alter table TB_WITH_DATA drop column description_bak;

步骤五：验证

1) 表结构验证

DESC TB_WITHOUT_DATA
Name        Null Type          
----------- ---- ------------- 
ID               NUMBER        
NAME             VARCHAR2(100) 
DESCRIPTION      CLOB 

DESC TB_WITH_DATA
Name        Null Type          
----------- ---- ------------- 
ID               NUMBER        
NAME             VARCHAR2(100) 
DESCRIPTION      CLOB  

2) 数据验证

select * from TB_WITH_DATA;

        ID NAME                       DESCRIPTION                                     
---------- -------------------------- ------------------------------------------------
         1 David Louis                He is capable of resolving such kind of issue   
         2 German Noemi               She is very beatiful and charming               
         3 Oliver Queen               He is main actor in the Green Arrow             
         4 Mark Williams              He plays snooker very well                      
         5 Sita Rama Raju Kata        I do not know this guy                          
         6 Promethus                  This is a very nice movie                       

 6 rows selected 

方法三：此方法适合此列包括数据和此列不包括数据两种情况

在解说方法三之前，须要包表恢复到准备阶段，因为时间关系，直接通过drop然后re-create方法,脚本例如以下：

drop table TB_WITHOUT_DATA;
drop table TB_WITH_DATA;

create table TB_WITHOUT_DATA
(
  id NUMBER,
  name VARCHAR2(100),
  description VARCHAR2(2000)
);

create table TB_WITH_DATA
(
  id NUMBER,
  name VARCHAR2(100),
  description VARCHAR2(2000)
);

insert into TB_WITH_DATA VALUES (1,'David Louis','He is capable of resolving such kind of issue');
insert into TB_WITH_DATA VALUES (2,'German Noemi','She is very beatiful and charming');
insert into TB_WITH_DATA VALUES (3,'Oliver Queen','He is main actor in the Green Arrow');
insert into TB_WITH_DATA VALUES (4,'Mark Williams','He plays snooker very well');
insert into TB_WITH_DATA VALUES (5,'Sita Rama Raju Kata','I do not know this guy');
insert into TB_WITH_DATA VALUES (6,'Promethus','This is a very nice movie');
commit;

步骤一：重命名两张表

rename TB_WITHOUT_DATA to TB_WITHOUT_DATA_BAK;
rename TB_WITH_DATA to TB_WITH_DATA_BAK;

步骤二：创建两张新表（通过下面语句创建两张表）

create table TB_WITHOUT_DATA
as
select id, name, to_clob(description) description
from TB_WITHOUT_DATA_BAK;

create table TB_WITH_DATA
as
select id, name, to_clob(description) description
from TB_WITH_DATA_BAK;

表结构与数据验证：

desc TB_WITHOUT_DATA
Name        Null Type          
----------- ---- ------------- 
ID               NUMBER        
NAME             VARCHAR2(100) 
DESCRIPTION      CLOB

desc TB_WITH_DATA
Name        Null Type          
----------- ---- ------------- 
ID               NUMBER        
NAME             VARCHAR2(100) 
DESCRIPTION      CLOB 

select * from TB_WITH_DATA;

select * from TB_WITH_DATA;

        ID NAME                       DESCRIPTION                                     
---------- -------------------------- ------------------------------------------------
         1 David Louis                He is capable of resolving such kind of issue   
         2 German Noemi               She is very beatiful and charming               
         3 Oliver Queen               He is main actor in the Green Arrow             
         4 Mark Williams              He plays snooker very well                      
         5 Sita Rama Raju Kata        I do not know this guy                          
         6 Promethus                  This is a very nice movie                       

 6 rows selected 

步骤三：删除备份表：

DROP TABLE TB_WITHOUT_DATA_BAK;
DROP TABLE TB_WITH_DATA_BAK;

-----------------------------------------------------------------------------------------------------------------------------------------------------------

假设您们在尝试的过程中遇到什么问题或者我的代码有错误的地方。请给予指正。很感谢！

联系方式：david.louis.tian@outlook.com

版权＠：转载请标明出处。