SQL logic to fail a check if any of the related customers has failed - sql

Question

I have the requirement to flag the customers Y only when all the related customers have also passed the check.
below are the two tables:
relationship table :
customer_id related_customer
1 1
1 2
1 3
2 1
2 2
2 3
3 1
3 2
3 3
11 11
11 22
22 11
22 22
Check table
customer_id check_flag
1 y
2 y
3 n
11 y
22 y
I want output like below:
customer_id paas_fail_flag
1 n
2 n
3 n
11 y
22 y
output justification: since 1,2,3 are related customers and since one of them (3) has n in table 2 , so all the related customers should also have n.
11,22 are related customers and both have y in table 2.so in output both should have y.

Answer 1

You need to join relationship to check and use conditional aggregation:
SELECT r.customer_id,
COALESCE(MAX(CASE WHEN c.check_flag = 'n' THEN c.check_flag END), 'y') paas_fail_flag
FROM relationship r INNER JOIN "check" c
ON c.customer_id = r.related_customer
GROUP BY r.customer_id
ORDER BY r.customer_id
See the demo.

Answer 2

Something like this? Sample data in lines #1 - 40; query begins at line #41:
SQL> WITH
2 -- sample data
3 rel (customer_id, related_customer)
4 AS
5 (SELECT 1, 1 FROM DUAL
6 UNION ALL
7 SELECT 1, 2 FROM DUAL
8 UNION ALL
9 SELECT 1, 3 FROM DUAL
10 UNION ALL
11 SELECT 2, 1 FROM DUAL
12 UNION ALL
13 SELECT 2, 2 FROM DUAL
14 UNION ALL
15 SELECT 2, 3 FROM DUAL
16 UNION ALL
17 SELECT 3, 1 FROM DUAL
18 UNION ALL
19 SELECT 3, 2 FROM DUAL
20 UNION ALL
21 SELECT 3, 3 FROM DUAL
22 UNION ALL
23 SELECT 11, 11 FROM DUAL
24 UNION ALL
25 SELECT 11, 22 FROM DUAL
26 UNION ALL
27 SELECT 22, 11 FROM DUAL
28 UNION ALL
29 SELECT 22, 22 FROM DUAL),
30 chk (customer_id, check_flag)
31 AS
32 (SELECT 1, 'y' FROM DUAL
33 UNION ALL
34 SELECT 2, 'y' FROM DUAL
35 UNION ALL
36 SELECT 3, 'n' FROM DUAL
37 UNION ALL
38 SELECT 11, 'y' FROM DUAL
39 UNION ALL
40 SELECT 22, 'y' FROM DUAL),
41 temp
42 AS
43 -- minimum CHECK_FLAG per customer and related customer
44 ( SELECT r.customer_id, r.related_customer, MIN (c.check_flag) mcf
45 FROM rel r JOIN chk c ON c.customer_id = r.related_customer
46 GROUP BY r.customer_id, r.related_customer)
47 SELECT customer_id, MIN (mcf) flag
48 FROM temp
49 GROUP BY customer_id
50 ORDER BY customer_id;
CUSTOMER_ID FLAG
----------- ----
1 n
2 n
3 n
11 y
22 y
SQL>

Answer 3

Assuming that your relationship data could be sparse, for example:
CREATE TABLE relationship ( customer_id, related_customer ) AS
SELECT 2, 3 FROM DUAL UNION ALL
SELECT 3, 1 FROM DUAL UNION ALL
SELECT 3, 2 FROM DUAL UNION ALL
SELECT 11, 22 FROM DUAL;
CREATE TABLE "CHECK" ( customer_id, check_flag ) AS
SELECT 1, 'y' FROM DUAL UNION ALL
SELECT 2, 'y' FROM DUAL UNION ALL
SELECT 3, 'n' FROM DUAL UNION ALL
SELECT 11, 'y' FROM DUAL UNION ALL
SELECT 22, 'y' FROM DUAL;
(Note: The below query will also work on your dense data, where every relationship combination is enumerated.)
Then you can use a hierarchical query:
SELECT customer_id,
MIN(check_flag) AS check_flag
FROM (
SELECT CONNECT_BY_ROOT(c.customer_id) AS customer_id,
c.check_flag AS check_flag
FROM "CHECK" c
LEFT OUTER JOIN relationship r
ON (r.customer_id = c.customer_id)
WHERE CONNECT_BY_ISLEAF = 1
CONNECT BY NOCYCLE
( PRIOR r.related_customer = c.customer_id
OR PRIOR c.customer_id = r.related_customer )
AND PRIOR c.check_flag = 'y'
)
GROUP BY
customer_id
ORDER BY
customer_id
Which outputs:
CUSTOMER_ID
CHECK_FLAG
1
n
2
n
3
n
11
y
22
y
db<>fiddle here