联表过滤MySQL中的信息

Question

I have the following question:

Input: 
Insurance table:
+-----+----------+----------+-----+-----+
| pid | tiv_2015 | tiv_2016 | lat | lon |
+-----+----------+----------+-----+-----+
| 1   | 10       | 5        | 10  | 10  |
| 2   | 20       | 20       | 20  | 20  |
| 3   | 10       | 30       | 20  | 20  |
| 4   | 10       | 40       | 40  | 40  |
+-----+----------+----------+-----+-----+
Output: 
+----------+
| tiv_2016 |
+----------+
| 45.00    |
+----------+

Explanation: The first record in the table, like the last record, meets both of the two criteria. The tiv_2015 value 10 is the same as the third and fourth records, and its location is unique.

The second record does not meet any of the two criteria. Its tiv_2015 is not like any other policyholders and its location is the same as the third record, which makes the third record fail, too. So, the result is the sum of tiv_2016 of the first and last record, which is 45.

My SQL query:

WITH tb1 AS(
SELECT DISTINCT a.pid,a.tiv_2016
FROM Insurance a
JOIN Insurance b
ON a.pid != b.pid AND a.tiv_2015 = b.tiv_2015
JOIN Insurance c
ON a.pid != c.pid AND (a.lat, a.lon) != (c.lat, c.lon)
)
SELECT SUM(tiv_2016) AS tiv_2016
FROM tb1

However, for the following input, my code failed to filter on the lat and lon columns:

| pid | tiv_2015 | tiv_2016 | lat | lon |
| --- | -------- | -------- | --- | --- |
| 1   | 10       | 5        | 10  | 10  |
| 2   | 20       | 20       | 20  | 20  |
| 3   | 10       | 30       | 20  | 20  |
| 4   | 10       | 40       | 40  | 40  |

My output:

| tiv_2016 |
| -------- |
| 75       |

The desired output:

| tiv_2016 |
| -------- |
| 45       |

When constructing tb1, my query also selected the row with pid = 3, which led to a total of 75.

I figured out other ways to return the expected output but just don't understand why my original SQL query won't work. Why does the ON a.pid.= c.pid AND (a,lat. a.lon),= (c.lat? c.lon) seem to do nothing?

Answer 1

I would first do an outer join of the Insurance table with itself to find all rows with unique lat/lon values (subquery sq below). Then I would do a sum of the tiv_2016 column grouped by the tiv_2015 column:

Schema (MySQL v5.7)

CREATE TABLE Insurance
    (`pid` int, `tiv_2015` int, `tiv_2016` int, `lat` int, `lon` int)
;
    
INSERT INTO Insurance
    (`pid`, `tiv_2015`, `tiv_2016`, `lat`, `lon`)
VALUES
    (1, 10, 5, 10, 10),
    (2, 20, 20, 20, 20),
    (3, 10, 30, 20, 20),
    (4, 10, 40, 40, 40)
;

---

Query #1

select sum(tiv_2016) as tiv_2016 from (
  select i1.* from Insurance i1
  left join Insurance i2 on i1.pid <> i2.pid and i1.lat = i2.lat and i1.lon = i2.lon
  where i2.pid is null
) sq
group by tiv_2015
;

tiv_2016
45

---

View on DB Fiddle