SQL：仅选择第二列的最常见值为 CERTAIN VALUE 的 id（按 id 分组）

Question

我在 SQL 中有一个表（使用 postgres），其中包含两列： videoid ，分类

videoid | classification
1       |20
1       |24
1       |24
1       |24
1       |24
2       |20
2       |20
2       |20
2       |20
2       |24
2       |24
3       |24
3       |24

我正在尝试检索最常见分类为 24 的所有 videoid。（答案应该只有 videoid 1 和 3） 当我使用查询时： （来自How to select most frequent value in a column per each id group?）

    SELECT DISTINCT ON (videoid) videoid, most_frequent_species FROM (
  SELECT videoid, classification AS most_frequent_species, count(*) as _count
FROM userviewed
GROUP BY videoid, classification) a
ORDER BY videoid, _count DESC
;

我检索结果集：

videoid     | most_frequent_species
    1       |24
    2       |20
    3       |24

但是当我尝试添加 WHERE 子句时：

WHERE classification = 24

我明白了：

videoid| most_frequent_species
1      |24
2      |24
3      |24

如何创建只检索的查询

videoid | most_frequent_species
1       |24
3       |24

Answer 1

您可以使用having 子句（本质上是一个post-group where 子句）和mode 函数：

select
    videoid
from
    userviewed
group by
    videoid
having
    mode() within group (order by classification) = 24

• mode docs
• having docs

Answer 2

一种方法使用distinct on然后过滤：

select *
from (select distinct on (videoid) videoid, category, count(*)
      from userviewed
      group by videoid, category
      order by videoid, count(*) desc
     ) vc
where category = 24;