查询返回每列中每个值的不同计数

Question

I'm looking to query a table structured like so

A B C D E
1 1 1 1 1
2 2 1 1 2
2 2 2 1 3
4 4 5 5 5

to see the column_name, distinct_value_in_column, count_of_distinct_value like so

A 1 1
A 2 2
A 4 2
B 1 1
B 2 2
B 4 1
C 1 2
C 2 1
C 5 1
D 1 3
D 5 1
E 1 1
E 2 1
E 3 1
E 5 1

Ultimately I'd like to sort by count descending, but it's getting to this structure that puzzles me. Can this be done in SQL generally? I'm using a postgresql instance if that has any impact.

In pseudo code I'm envisioning

select column_names, distinct a, b, c, d, e, count (`distinct a, b, c, d, e) as ct
from table
group by column_names
order by ct desc

Answer 1

Sure. Stack the data first using UNION .

select
    col,
    val,
    count(*)
from
    (
        select
            'A' as col,
            a as val
        from
            table
        union all select
            'B' as col,
            b as val
        from
            table
        union all select
            'C' as col,
            c as val
        from
            table
        union all select
            'D' as col,
            d as val
        from
            table
        union all select
            'E' as col,
            e as val
        from
            table
    )
group by
    col,
    val
order by
    count(*) desc;

Answer 2

You can use the following query:

SELECT col, val, count(*)
FROM (
  SELECT unnest(array['A','B','C','D','E']) AS col,
         unnest(array[A,B,C,D,E]) AS val
  FROM mytable) AS t
GROUP BY col, val

Demo here