准备语句：DROP TABLE ix_test;

CREATE TABLE ix_test

(id_1 varchar(20) NOT NULL,

PRIMARY KEY(id_1));

INSERT INTO ix_test VALUES (1),(2),(3),(4),(5),(6),(7),(8),(9),(10),(11);

mysql会将数字在×××和字符串之间自动转换！

这样下面两条语句的结果是一样的：

SELECT * FROM ix_test WHERE id_1=1;

SELECT * FROM ix_test WHERE id_1='1';

但是在索引使用情况方面，结果就完全不一样了！第一条不使用索引，第二条使用索引！

两条语句，第一条使用了索引，但是扫描了全表，第二条直接索引到数据，只需要读取一行！mysql> explain select * from ix_test where id_1=1;

+----+-------------+---------+-------+---------------+---------+---------+------+------+--------------------------+

| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |

+----+-------------+---------+-------+---------------+---------+---------+------+------+--------------------------+

| 1 | SIMPLE | ix_test | index | PRIMARY | PRIMARY | 302 | NULL | 11 | Using where; Using index |

+----+-------------+---------+-------+---------------+---------+---------+------+------+--------------------------+

1 row in set (0.00 sec)

mysql> explain select * from ix_test where id_1='1';

+----+-------------+---------+-------+---------------+---------+---------+-------+------+-------------+

| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |

+----+-------------+---------+-------+---------------+---------+---------+-------+------+-------------+

| 1 | SIMPLE | ix_test | const | PRIMARY | PRIMARY | 302 | const | 1 | Using index |

+----+-------------+---------+-------+---------------+---------+---------+-------+------+-------------+

1 row in set (0.01 sec)

但是如果将id_1字段变为×××，后面用×××或者是字符串去匹配都可以使用索引，而且索引直接命中！

所以，多么坑爹的mysql sql优化器，多么痛的领悟！

得出结论，对于where后面字段类型为字符串的数字，如果用×××去匹配(就是不用引号引上数字)，则不能由索引直接命中，需要全部扫描。

特别需要注意！