都说Oracle存在NUMBER和VARCHAR2类型的隐式转换,严格意义上需要避免,但为何需要避免,从下面的实验进行验证。
1. 创建测试表和索引
create table tn (id number, name varchar2(1));
create index idx_tn on tn (id);
create index idx_tn on tn (name);
分别对NUMBER类型的id字段,VARCHAR2类型的name字段创建索引。
2. 查看VARCHAR2->NUMBER的隐式转换
SQL> select * from tn where id = 1;
no rows selected
Execution Plan
----------------------------------------------------------
Plan hash value: 3532270966
---------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
---------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 13 | 1 (0)| 00:00:01 |
|* 1 | INDEX RANGE SCAN| IDX_TN | 1 | 13 | 1 (0)| 00:00:01 |
---------------------------------------------------------------------------
"where id = 1"用的是列索引范围扫描。
SQL> select * from tn where id = ‘123‘;
no rows selected
Execution Plan
----------------------------------------------------------
Plan hash value: 3532270966
---------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
---------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 13 | 1 (0)| 00:00:01 |
|* 1 | INDEX RANGE SCAN| IDX_TN | 1 | 13 | 1 (0)| 00:00:01 |
---------------------------------------------------------------------------
"where id = ‘123‘",Oracle会将字符类型的123转换为NUMBER类型进行比较,此处仍可使用索引范围扫描,说明VARCHAR2->NUMBER的隐式转换,未对索引产生影响。
3. 查看NUMBER->VARCHAR2的隐式转换
SQL> select * from tn where name = ‘123‘;
no rows selected
Execution Plan
----------------------------------------------------------
Plan hash value: 479240418
-------------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
-------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 15 | 1 (0)| 00:00:01 |
| 1 | TABLE ACCESS BY INDEX ROWID| TN | 1 | 15 | 1 (0)| 00:00:01 |
|* 2 | INDEX RANGE SCAN | IDX_TN_NAME | 1 | | 1 (0)| 00:00:01 |
-------------------------------------------------------------------------------------------
"where name = ‘123‘"使用的是索引范围扫描。
SQL> select * from tn where name = 123;
no rows selected
Execution Plan
----------------------------------------------------------
Plan hash value: 2655062619
--------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 15 | 2 (0)| 00:00:01 |
|* 1 | TABLE ACCESS FULL| TN | 1 | 15 | 2 (0)| 00:00:01 |
--------------------------------------------------------------------------
"where name = 123",Oracle会将数值类型的123转换为VARCHAR2字符类型,和name进行比较,此处用了全表扫描,说明name的列索引失效。
总结:
1. NAME和VARCHAR2之间可以进行隐式转换,其中VARCHAR2->NUMBER不会导致索引失效,NUMBER->VARCHAR2会让索引失效,因此这种隐式转换,是需要注意避免。
2. 之所以VARCHAR2->NUMBER不会让索引失效,我猜测是转换为where id = to_number(‘123‘)。NUMBER->VARCHAR2会让索引失效,我猜测是转换为where to_number(name) = 123。
3. 引申知识点,之所以上面id和name使用的是索引范围扫描,是因为建立的是非唯一B树索引,如果是unique索引,则会使用UNIQUE INDEX SCAN的扫描方式。