是否可以在子查询中返回多于一行

Question

My Query

SELECT c.COLUMN_A,
(select count(b.COLUMN_B) from SAME_TABLE_NAME b where COLUMN_B='X' GROUP by   COLUMN_A) as ALIAS_NAME 

FROM SAME_TABLE_NAME c

Above query throw Error as

single-row sub query returns more than one row.

but i need all the row it is returning

Answer 1

Try:

    SELECT COLUMN_A, COUNT(COLUMN_B)  AS ALIAS_NAME 
      FROM SAME_TABLE_NAME 
      WHERE COLUMN_B='X' GROUP BY COLUMN_A

If that's not right, please post your real schema and an example of your data.

Answer 2

you are getting multiple results in your column. your looking for something like this.

select  COLUMN_A, count(b.COLUMN_B) from SAME_TABLE_NAME b 
            where COLUMN_B='X' GROUP by COLUMN_A

Answer 3

You may try this.

SELECT c.COLUMN_A,
(select count(b.COLUMN_B) from SAME_TABLE_NAME b 
 where COLUMN_B='X' 
 AND C.COLUMN_A=B.COLUMN_A   --- THIS WHERE CLAUSE IS ADDED TO MAP ALL YOUR COUNT RECORD WITH ITS RESPECTED COLUMN
 GROUP by B.COLUMN_A) as ALIAS_NAME 
FROM SAME_TABLE_NAME c

Answer 4

A scalar subquery should return at most one row. Hence, GROUP BY is not appropriate. If you want to express this using a subquery, you want a correlated subquery without GROUP BY :

SELECT c.COLUMN_A,
       (SELECT count(b.COLUMN_B) 
        FROM SAME_TABLE_NAME b 
        WHERE c.COLUMN_A = b.COLUMN_A AND
              b.COLUMN_B = 'X' 
       ) as ALIAS_NAME 
FROM SAME_TABLE_NAME c;

However, the better approach is to use window functions, available in the more recent versions of MySQL:

SELECT c.COLUMN_A,
       SUM(c.COLUMN_B = 'X') OVER (PARTITION BY c.COLUMN_A) as ALIAS_NAME 
FROM SAME_TABLE_NAME c;

Note that you should qualify all column references. This is especially true with subqueries, where the scoping rules may sometimes not do what you intend.