行转列,列转行这种需求,在 DBA 的从业生涯中过分常见。今天,就结合小伙伴儿的实际需求,跟大家分享一个行转列的实际案例。
现存在表A:table_org,表结构简略抽象如下:
+------------------+---------------------+------+-----+-------------------+-----------------------------+
| Field | Type | Null | Key | Default | Extra |
+------------------+---------------------+------+-----+-------------------+-----------------------------+
| id | int(10) unsigned | NO | PRI | NULL | auto_increment |
| uri | varchar(4096) | NO | | | |
| ... | ... | ... | | ... | |
| insert_time | datetime | NO | | CURRENT_TIMESTAMP | |
| update_time | datetime | NO | | CURRENT_TIMESTAMP | on update CURRENT_TIMESTAMP |
+------------------+---------------------+------+-----+-------------------+-----------------------------+
现在存在的问题是,由于设计问题,uri 字段中可能存在多个值,比如:
/daily/list,/daily/detail,/daily/info,....
经查询,最多可能同时存在15个 uri,但是每行 uri 个数不固定,可能是 0 也有可能是 2,3,5... 查询方法:
mysql> select distinct(length(uri) - length((replace(uri, ',', ''))))+1 from tbl_org;
+---------------------------------------------------+
| (length(uri) - length((replace(uri, ',', ''))))+1 |
+---------------------------------------------------+
| 1 |
| 4 |
| 3 |
| 2 |
| 14 |
| 7 |
| 13 |
| 5 |
| 6 |
| 10 |
| 8 |
| 9 |
+---------------------------------------------------+
13 rows in set (0.03 sec)
说明:我们将',' 替换为 '',损失的长度即为',' 的个数,uri 的个数= ',' 的个数+1
行转列的一个常用方法就是:join
我们在 test 库中创建一个 id 为 1-15 的表:(也可以更大点儿,比如:1-100)
mysql> select * from test.tina_num;
+----+
| id |
+----+
| 1 |
| 2 |
| 3 |
| 4 |
| 5 |
| 6 |
| 7 |
| 8 |
| 9 |
| 10 |
| 11 |
| 12 |
| 13 |
| 14 |
| 15 |
+----+
15 rows in set (0.03 sec)
使用语句:
select substring_index(substring_index(uri,',', t.id),',',-1) from tbl_org d join tina_num t on 1=1 where t.id <= (length(uri)-length(replace(uri,',','')) + 1);
即可得到结果:
+--------------------------------------------------------+
| substring_index(substring_index(uri,',', t.id),',',-1) |
+--------------------------------------------------------+
| /daily/list |
| /daily/detail |
| /daily/info |
| ... |
+--------------------------------------------------------+
说明:
1.获得改行该字段中 uri 的个数, 我们将此记为:row_num
(length(uri)-length(replace(uri,',','')) + 1)
2.有 row_num 个 uri,就应该转化为 row_num 行,就要和 tina_num 表中的 row_num 行做 join
:
from tbl_org d join tina_num t on 1=1 where t.id <= row_num
组合起来就是:
from tbl_org d join tina_num t on 1=1 where t.id <= (length(uri)-length(replace(uri,',','')) + 1)
3.使用 substring_index 将 uri 切段儿:
substring_index(uri,',', t.id)
MySQL官方文档中对substring_index
的定义为:返回指定个数切片之前的所有字符
Return a substring from a string before the specified number of occurrences of the delimiter
简单示例:
mysql> select substring_index('A,B', ',', 1);
+--------------------------------+
| substring_index('A,B', ',', 1) |
+--------------------------------+
| A |
+--------------------------------+
1 row in set (0.02 sec)
mysql> select substring_index('A,B', ',', 2);
+--------------------------------+
| substring_index('A,B', ',', 2) |
+--------------------------------+
| A,B |
+--------------------------------+
1 row in set (0.02 sec)
所以逐次获得1,2,...n 个 uri 的语句就是:
select substring_index(substring_index(uri,',', t.id),',',-1)
4.合起来的整条语句:
select substring_index(substring_index(uri,',', t.id),',',-1) from tbl_org d join tina_num t on 1=1 where t.id <= (length(uri)-length(replace(uri,',','')) + 1);