在一个查询中组合多列

Question

Table A

+------------+--------------+--------------+
+  ID        +  Name        +   Company ID +
+------------+--------------+--------------+
+   1        +  Steve       +      1       +
+   2        +  Mile        +      2       +
+   3        +  Steve       +      2       +
+   4        +  Del Piero   +      2       +
+   5        +  Jack        +              +
============================+==============+

Table B

+------------+--------------+
+  ID        +  Company     +
+------------+--------------+
+   1        +  A           +
+   2        +  B           +
============================+

What I want

+----------+------------+----------------+
+ # of id  +  # OF Name +  # of B Company+
+----------+------------+----------------+
+ 5        +  4         +  3             +
+========================================+

I have tried Union//Union All

SELECT count(id), count(name) FROM table A
UNION 
SELECT count(company) FROM table B where company = 'B'

I should keep these columns but Union doesn't allow a different number of columns.

Any ideas would be appreciated

Answer 1

You can join the tables to recover the company name associated to each record in tablea , then aggregate:

select
    count(*) no_ids,
    count(distinct a.name) no_names,
    sum(b.comany = 'B') no_b_company
from tablea a
inner join tableb b on b.id = a.company_id

Presumably, id is the primary key of tablea , so we just use count(*) to summarize this column. no_names counts how many different names can be seen in the corresponding column, and sum(b.company = 'B') increase by 1 everytime company B is seen.

If there is a possibility of missing companies in tableb , you can change the inner join to a left join .

Answer 2

Try this one

SELECT count(id),
count(name),
(SELECT count(company) FROM table B where company)
FROM table A