\'Group by\' 没有按照我需要的顺序进行分组

Question

我想对啤酒进行“分组”，以便将它们组合在一起，并在单独的列中列出尊重的总评分和品尝者（评论啤酒的人）。

这是我的代码：

create or replace view tasters_avg_ratings1
as
select a.taster as taster, a.beer as beer, round(avg(a.rating),1) as rating 
from allratings a
group by beer, taster
;

然而我的输出看起来像这样：

beers=# select * from tasters_avg_ratings1;
 taster |          beer          | rating 
--------+------------------------+--------
 Peter  | XXXX                   |    5.0
 Sarah  | James Squire Pilsener  |    3.0
 Raghu  | Sierra Nevada Pale Ale |    3.0
 Hector | Fosters                |    3.0
 John   | Chimay Red             |    3.0
 John   | Sierra Nevada Pale Ale |    5.0
 Geoff  | James Squire Pilsener  |    4.0
 Ramez  | Sierra Nevada Pale Ale |    4.0
 John   | 80/-                   |    4.0
 John   | Rasputin               |    4.0
 Adam   | Old                    |    4.0
 John   | Crown Lager            |    2.0
 Jeff   | Sierra Nevada Pale Ale |    4.0
 Sarah  | Burragorang Bock       |    4.0
 Sarah  | Scharer's Lager        |    3.0
 Sarah  | New                    |    2.0
 Geoff  | Redback                |    4.0
 Adam   | Victoria Bitter        |    1.0
 Sarah  | Victoria Bitter        |    1.0
 Raghu  | Rasputin               |    3.0
 Ramez  | Bigfoot Barley Wine    |    3.0
 Hector | Sierra Nevada Pale Ale |    4.0
 Sarah  | Old                    |    3.0
 Jeff   | Burragorang Bock       |    3.0
 John   | Empire                 |    3.0
 Sarah  | James Squire Amber Ale |    3.0
 Rose   | Redback                |    5.0
 Geoff  | Empire                 |    3.0
 Adam   | New                    |    1.0
 Jeff   | Rasputin               |    1.0
 Raghu  | Old Tire               |    5.0
 John   | Victoria Bitter        |    1.0
(32 rows)

如您所见，啤酒并未分组在一起。例如，理想情况下，“Victoria Bitter”啤酒应该作为一组展示，而不是分开展示。

期望的结果是使用“order by”实现的。例如：

create or replace view tasters_avg_ratings1
as
select a.taster as taster, a.beer as beer, round(avg(a.rating),1) as rating 
from allratings a
group by beer, taster
order by a.beer
;

输出：

beers=# select * from tasters_avg_ratings1;
 taster |          beer          | rating 
--------+------------------------+--------
 John   | 80/-                   |    4.0
 Ramez  | Bigfoot Barley Wine    |    3.0
 Jeff   | Burragorang Bock       |    3.0
 Sarah  | Burragorang Bock       |    4.0
 John   | Chimay Red             |    3.0
 John   | Crown Lager            |    2.0
 Geoff  | Empire                 |    3.0
 John   | Empire                 |    3.0
 Hector | Fosters                |    3.0
 Sarah  | James Squire Amber Ale |    3.0
 Geoff  | James Squire Pilsener  |    4.0
 Sarah  | James Squire Pilsener  |    3.0
 Adam   | New                    |    1.0
 Sarah  | New                    |    2.0
 Adam   | Old                    |    4.0
 Sarah  | Old                    |    3.0
 Raghu  | Old Tire               |    5.0
 Jeff   | Rasputin               |    1.0
 John   | Rasputin               |    4.0
 Raghu  | Rasputin               |    3.0
 Geoff  | Redback                |    4.0
 Rose   | Redback                |    5.0
 Sarah  | Scharer's Lager        |    3.0
 Hector | Sierra Nevada Pale Ale |    4.0
 Jeff   | Sierra Nevada Pale Ale |    4.0
 John   | Sierra Nevada Pale Ale |    5.0
 Raghu  | Sierra Nevada Pale Ale |    3.0
 Ramez  | Sierra Nevada Pale Ale |    4.0
 Adam   | Victoria Bitter        |    1.0
 John   | Victoria Bitter        |    1.0
 Sarah  | Victoria Bitter        |    1.0
 Peter  | XXXX                   |    5.0
(32 rows)

因此，虽然我知道 order by 可以达到我的结果，但为什么“group by”不做同样的事情？这是令人沮丧的，因为我在 Internet 上看到过许多使用“分组依据”的示例，并且特别是在与我类似的非聚合列和聚合列的情况下。例如：https://learnsql.com/blog/error-with-group-by/，提示 #3。

任何帮助将不胜感激，谢谢！

Answer 1

GROUP BY 仅用于计算聚合值（在本例中为平均评分）。它与结果显示时的顺序无关。正如您所提到的，您需要使用 ORDER BY 来获得所需的顺序。