mysql 按相似度来排序
假设有这几条记录
1：刘世允
2：郑胜浩
3：申东烨
4: 申东烨,郑胜浩,刘世允,安英美
5: 郑胜浩,刘世允,安英美
6: 申东烨,郑胜浩
7：刘世允,安英美
给出的查询tag
$tag = '申东烨,郑胜浩,刘世允,安英美,李尚勋';
按照相同词的个数排序
查询的结果
1: 申东烨,郑胜浩,刘世允,安英美
2: 郑胜浩,刘世允,安英美
3: 申东烨,郑胜浩
4：刘世允,安英美
5：刘世允
6：郑胜浩
7：申东烨
回复内容：
数据库tag字段内容格式：a,b,c,d ...
假设有这几条记录
1：刘世允
2：郑胜浩
3：申东烨
4: 申东烨,郑胜浩,刘世允,安英美
5: 郑胜浩,刘世允,安英美
6: 申东烨,郑胜浩
7：刘世允,安英美
给出的查询tag
$tag = '申东烨,郑胜浩,刘世允,安英美,李尚勋';
按照相同词的个数排序
查询的结果
1: 申东烨,郑胜浩,刘世允,安英美
2: 郑胜浩,刘世允,安英美
3: 申东烨,郑胜浩
4：刘世允,安英美
5：刘世允
6：郑胜浩
7：申东烨
----------------建表
mysql> create table testsort (tag varchar(100) charset "GBK");
Query OK, 0 rows affected (0.51 sec)
```sql
mysql> describe testsort
-> ;
+-------+--------------+------+-----+---------+-------+
| Field | Type | Null | Key | Default | Extra |
+-------+--------------+------+-----+---------+-------+
| tag | varchar(100) | YES | | NULL | |
+-------+--------------+------+-----+---------+-------+
1 row in set (0.07 sec)
----------------插入数据
```sql
mysql> insert into testsort(id,tag) values(1,"刘世允"),
-> (2,"郑胜浩"),
-> (3,"申东烨"),
-> (4,"申东烨,郑胜浩,刘世允,安英美"),
-> (5,"郑胜浩,刘世允,安英美"),
-> (6,"申东烨,郑胜浩"),
-> (7,"刘世允,安英美");
Query OK, 7 rows affected (0.07 sec)
Records: 7 Duplicates: 0 Warnings: 0
----------------方法
mysql> select id,
-> tag,
-> char_length(tag)-char_length(replace(tag,',',''))+1 as cnt
-> from testsort;
+------+-----------------------------------------+------+
| id | tag | cnt |
+------+-----------------------------------------+------+
| 1 | 刘世允 | 1 |
| 2 | 郑胜浩 | 1 |
| 3 | 申东烨 | 1 |
| 4 | 申东烨,郑胜浩,刘世允,安英美 | 4 |
| 5 | 郑胜浩,刘世允,安英美 | 3 |
| 6 | 申东烨,郑胜浩 | 2 |
| 7 | 刘世允,安英美 | 2 |
+------+-----------------------------------------+------+
7 rows in set (0.00 sec)
----------------另外注意
Responses below will get you there. However, don't forget to use CHAR_LENGTH() instead of LENGTH() if you're using multibyte characters. – inhan Sep 10 '12 at 3:03
http://*.com/questions/12344795/count-the-number-of-occurences-of-a-string-in-a-varchar-field