如何在mysql中将列拆分为多个列

时间:2015-03-17 07:30:29

标签: mysql sql dynamic

我有这样的专栏


+--------------------------+
| Marks                    |
+--------------------------+
|Maths-80,Phy-100,Che-99   |
|Maths-90,Phy-60           | 
|Phy-82,Che-65             |
|Che-90                    |
|Maths-33,Phy-89,Che-65 | |Maths-50,Phy-43,Che-59 | +--------------------------+
我只给3个样本科目但是在真实数据中它可能是用逗号分隔的任意数量的科目。为此我想通过拆分标记列显示每个主题的标记,并将其显示在多个列中,如下面的示例所示。

 +-------------------------------------+-------------+------------+
| Marks                  | Maths       | Phy         | Che        |
+-------------------------------------+-------------+-------------+                               
Maths-80,Phy-100,Che-99  | 80          |   100      |     99      |
Maths-90,Phy-60          | 90          |   60       |     0       |
Phy-82,Che-65            |  0          |   82       |     65      |
Che-90                   | 90          |   0        |     0       |
Maths-33,Phy-89,Che-65   | 33          |   89       |     65      |
Maths-50,Phy-43,Che-59   | 50          |   43       |     59      |
+-------------------------------------+-------------+-------------+  

1 个答案:

答案 0 :(得分:2)

使用给定的示例数据,您可以执行以下操作 -

mysql> create table test (marks varchar(100));
Query OK, 0 rows affected (0.12 sec)

mysql> insert into test values ('Maths-80,Phy-100,Che-99'),('Maths-90,Phy-60'),('Phy-82,Che-65'),('Che-90'),('Maths-33,Phy-89,Che-65'),('Maths-50,Phy-43,Che-59');
Query OK, 6 rows affected (0.04 sec)
Records: 6  Duplicates: 0  Warnings: 0

mysql> select * from test ;
+-------------------------+
| marks                   |
+-------------------------+
| Maths-80,Phy-100,Che-99 |
| Maths-90,Phy-60         |
| Phy-82,Che-65           |
| Che-90                  |
| Maths-33,Phy-89,Che-65  |
| Maths-50,Phy-43,Che-59  |
+-------------------------+
6 rows in set (0.00 sec)

现在使用locatesubstring_index我们可以将值提取为

select marks, 
case 
 when locate('Maths',marks) > 0 then substring_index(substring_index(marks,'Maths-',-1),',',1) else 0 end 
as Maths ,
case 
 when locate('Phy',marks) > 0 then substring_index(substring_index(marks,'Phy-',-1),',',1) else 0 end 
as Phy ,
case 
 when locate('Che',marks) > 0 then substring_index(substring_index(marks,'Che-',-1),',',1) else 0 end 
as Che 
from test ;

输出:

+-------------------------+-------+------+------+
| marks                   | Maths | Phy  | Che  |
+-------------------------+-------+------+------+
| Maths-80,Phy-100,Che-99 | 80    | 100  | 99   |
| Maths-90,Phy-60         | 90    | 60   | 0    |
| Phy-82,Che-65           | 0     | 82   | 65   |
| Che-90                  | 0     | 0    | 90   |
| Maths-33,Phy-89,Che-65  | 33    | 89   | 65   |
| Maths-50,Phy-43,Che-59  | 50    | 43   | 59   |
+-------------------------+-------+------+------+
6 rows in set (0.00 sec)