mysql子查询先排序再分组的问题

有表如下，要找到 family_name 为 Green 和 Smith 的身高最高的两个人的 given_name。

+----+-------------+------------+--------+
| id | family_name | given_name | height |
+----+-------------+------------+--------+
|  1 | Smith       | Mike       |    157 |
|  2 | Smith       | Ashlin     |    165 |
|  3 | Smith       | Eason      |    181 |
|  4 | Smith       | Kaley      |    175 |
|  5 | Green       | Jack       |    164 |
|  6 | Green       | Kate       |    179 |
|  7 | Green       | Kathryn    |    168 |
|  8 | Green       | Waller     |    187 |
|  9 | Brown       | Mary       |    178 |
| 10 | Brown       | Lana       |    173 |
| 11 | Brown       | Lange      |    175 |
| 12 | Brown       | Randy      |    177 |
+----+-------------+------------+--------+

在子查询里先按身高排序，再在外面用 family_name 分组：

select given_name from (
	SELECT * from height 
	where family_name in ('Green','Smith') 
	order by height desc 
) b 
group by family_name;

结果外层的分组并未按照子查询内的顺序进行。

+------------+
| given_name |
+------------+
| Jack       |
| Mike       |
+------------+

查询效果和直接查结果一样。

select given_name from height 
where family_name in ('Green','Smith') 
group by family_name
order by height desc 

两句 sql explain 的结果也一样。

id	select_type	table	partitions	type	possible_keys	key	key_len	ref	rows	filtered	Extra
1	SIMPLE	height	NULL	ALL	NULL	NULL	NULL	NULL	12	20.00	Using where; Using temporary; Using filesort

但是在上述的子查询 sql 中使用 group by，就能达到预期结果。

select given_name from (
	SELECT * from height 
	where family_name in ('Green','Smith') 
	group by id
	order by height desc 
) b 
group by family_name;

+------------+
| given_name |
+------------+
| Waller     |
| Eason      |
+------------+

再看 该语句的 explain：

id	select_type	table	partitions	type	possible_keys	key	key_len	ref	rows	filtered	Extra
1	PRIMARY	<derived2>	NULL	ALL	NULL	NULL	NULL	NULL	2	100.00	Using temporary; Using filesort
2	DERIVED	height	NULL	ALL	PRIMARY	NULL	NULL	NULL	12	20.00	Using where; Using filesort

mysql 5.7 版本对子查询进行了优化，子查询会被改为联结查询的形式以节省时间和资源，除非子查询内包含union 、group by、distinct、limit 或使用了聚合函数。
具体原因查看https://blog.csdn.net/luis_ora/article/details/80178931