SQL-搜索两个表的sum（amount）列不相等的情况

Question

Table1 has one row per id, table 2 has one row per id where entrytype = 0 and multiple rows per id where entrytype = 1.

I run the following queries:

SELECT sum(amount) FROM table1 where id = 'AUS|License|Maintenance|Aug_2016'

select sum(amount) from table2 where entrytype = 0 and id = 'AUS|License|Maintenance|Aug_2016'

select sum(amount) from table2 where entrytype = 1 and id = 'AUS|License|Maintenance|Aug_2016'

And get the following results:

7689.12

7689.12

7689.119999

Now, I want to search both tables' sum(amount) for cases where table1 sum(amount) <> table2 sum(amount) for a particular id

Answer 1

If you want it for a particular id, you can use cross join to bring them together:

select t1.amount, t2_0.amount, t2_1.amount
from (SELECT sum(amount) as amount FROM table1 where id = 'AUS|License|Maintenance|Aug_2016'
     ) t1 cross join
     (select sum(amount) as amount from table2 where entrytype = 0 and id = 'AUS|License|Maintenance|Aug_2016'
     ) t2_0 cross join
     (select sum(amount) as amount from table2 where entrytype = 1 and id = 'AUS|License|Maintenance|Aug_2016'
     ) t2_1;

Answer 2

If I've understood your requirement correctly, you could join each of your queries on their common id field, and then simply select those id values for which the summation is not equal across the three subqueries, eg:

SELECT a.id, a.s1, b.s2, c.s3
FROM
(
    (
        SELECT id, SUM(amount) AS s1
        FROM table1 
        GROUP BY id
    ) a
    INNER JOIN
    (
        SELECT id, SUM(amount) AS s2
        FROM table2
        WHERE entrytype = 0
        GROUP BY id
    ) b ON a.id = b.id
)
INNER JOIN
(
    SELECT id, SUM(amount) AS s3
    FROM table2
    WHERE entrytype = 1
    GROUP BY id
) c ON a.id = c.id
WHERE NOT (a.s1 = b.s2 AND a.s1 = c.s3)

You didn't specify a DBMS, so I've included the extra parentheses required by MS Access, just in case.

By using INNER JOINs , I've also assumed that you're only interested in records which exist in all three subqueries; if one of the subqueries is the master, join the other two subqueries using LEFT JOINs so that all records in the master are returned.

---

Including the two additional columns mentioned in the subsequent comments -

SELECT a.id, a.s1, b.s2, c.s3, a.s1 - b.s2 AS s4, a.s1 - c.s3 AS s5
FROM
(
    (
        SELECT id, SUM(amount) AS s1
        FROM table1 
        GROUP BY id
    ) a
    INNER JOIN
    (
        SELECT id, SUM(amount) AS s2
        FROM table2
        WHERE entrytype = 0
        GROUP BY id
    ) b ON a.id = b.id
)
INNER JOIN
(
    SELECT id, SUM(amount) AS s3
    FROM table2
    WHERE entrytype = 1
    GROUP BY id
) c ON a.id = c.id
WHERE NOT (a.s1 = b.s2 AND a.s1 = c.s3)