在SELECT子句之前或之后解析GROUP BY子句

Question

我正在尝试理解MySQL DBMS中SELECT查询执行的逻辑顺序。

阅读this和this个问题，订单可以是：

  

  
1. FROM
  
2. ON
  
3. JOIN
  
4. WHERE
  
5. GROUP BY
  
6. WITH CUBE或WITH ROLLUP
  
7. HAVING
  
8. 选择
  
9. DISTINCT
  
10. ORDER BY
  
11. TOP
  

或者：

  

  
1. FROM clause
  
2. WHERE子句
  
3. SELECT子句
  
4. GROUP BY子句
  
5. HAVING条款
  
6. ORDER BY子句
  

正如你所看到的那样，两者的顺序相反，所以我开始摆弄MySQL。

表格（PROGETTO）如下：

+----------+-------------+------+-----+---------+-------+
| Field    | Type        | Null | Key | Default | Extra |
+----------+-------------+------+-----+---------+-------+
| NOME_P   | varchar(15) | NO   | UNI | NULL    |       |
| NUMERO_P | int(11)     | NO   | PRI | NULL    |       |
| SEDE_P   | varchar(15) | YES  |     | NULL    |       |
| NUM_D    | int(11)     | NO   | MUL | NULL    |       |
+----------+-------------+------+-----+---------+-------+

数据集是：

+-----------------+----------+-----------+-------+
| NOME_P          | NUMERO_P | SEDE_P    | NUM_D |
+-----------------+----------+-----------+-------+
| ProdottoX       |        1 | Bellaire  |     5 |
| ProdottoY       |        2 | Sugarland |     5 |
| ProdottoZ       |        3 | Houston   |     5 |
| Informatizzazio |       10 | Stafford  |     4 |
| Riorganizzazion |       20 | Houston   |     1 |
| Nuove opportuni |       30 | Stafford  |     4 |
+-----------------+----------+-----------+-------+

所以，我发起了第一个查询：

mysql> SELECT NOME_P FROM PROGETTO GROUP BY NUM_D;
+-----------------+
| NOME_P          |
+-----------------+
| Riorganizzazion |
| Informatizzazio |
| ProdottoX       |
+-----------------+

我想，“好的顺序必须是第一个，但让我们再试一次......”然后我执行了：

mysql> SELECT NOME_P AS NNN FROM PROGETTO GROUP BY  NNN;
+-----------------+
| NNN             |
+-----------------+
| Informatizzazio |
| Nuove opportuni |
| ProdottoX       |
| ProdottoY       |
| ProdottoZ       |
| Riorganizzazion |
+-----------------+

......这让我很困惑。