sql find_in_set 和 like的区别

mysql> select id, backup_time from backup_policy limit 2;
+----+-------------+
| id | backup_time |
+----+-------------+
| 21 | Wen,Sun     |
| 24 |             |
+----+-------------+
2 rows in set (0.00 sec)
mysql> select count(*) from backup_policy where find_in_set('Wen', backup_time);
+----------+
| count(*) |
+----------+
|        2 |
+----------+
1 row in set (0.00 sec)

mysql> select count(*) from backup_policy where backup_time like '%Wen%';
+----------+
| count(*) |
+----------+
|        2 |
+----------+
1 row in set (0.00 sec)

find_in_set函数是查询某个字段（以英文逗号隔开的字符串）中是否包含某个值。

like是比较常用的模糊查询，是把字段整体看成是查询对象。

find_in_set 是精确匹配，字段值以英文 ‘,’ 分隔，而like 是广泛的模糊查询，字段中无需使用分隔符。

两者表示的意义、用法、逻辑是不同的，作用是否相同要看你要查询的结果了。

我想用explain来测试下两者的区别，但是，是失败的

mysql> EXPLAIN select count(*) from backup_policy where find_in_set('Wen', backup_time);
+----+-------------+---------------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table         | type | possible_keys | key  | key_len | ref  | rows | Extra       |
+----+-------------+---------------+------+---------------+------+---------+------+------+-------------+
|  1 | SIMPLE      | backup_policy | ALL  | NULL          | NULL | NULL    | NULL | 3169 | Using where |
+----+-------------+---------------+------+---------------+------+---------+------+------+-------------+
1 row in set (0.00 sec)

mysql> explain select count(*) from backup_policy where backup_time like '%Wen%';
+----+-------------+---------------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table         | type | possible_keys | key  | key_len | ref  | rows | Extra       |
+----+-------------+---------------+------+---------------+------+---------+------+------+-------------+
|  1 | SIMPLE      | backup_policy | ALL  | NULL          | NULL | NULL    | NULL | 3169 | Using where |
+----+-------------+---------------+------+---------------+------+---------+------+------+-------------+
1 row in set (0.00 sec)

两者的结果是一样的。。。。。。

至于两个sql语句的性能谁高谁低，我感觉这个跟数据库的表结构、索引、数据量等因素有关，我还没有大量数据去测试，这个作业留给大家吧！