How to make an inner join while maintaining unique rows - mysql

Question

I have a ternary relationship in which I stablish the relation between Offers, Profiles, and Skills. The ternary relationship table, called ternary for example, has the IDs of the three tables as primary key. It could look something like this:
id_Offer - id_Profile - id_Skill
1 - 1 - 1
1 - 1 - 2
1 - 1 - 3
1 - 2 - 1
2 - 1 - 1
2 - 3 - 2
2 - 1 - 3
2 - 5 - 1
[and so on, there would be more registers for each id_Offer from Offer but I want to limit the example]
So I have 2 offers in total, with a number of profiles in each one.
The table Offer looks something like this:
Offer - business_name
1 - business-1
2 - business-1
3 - business-1
4 - business-1
5 - business-2
6 - business-2
7 - business-2
8 - business-3
So when I do a query like
select distinct id_offer, business_name, COUNT(*)
FROM Offer
GROUP BY business_name
Order by COUNT(*);
I get that for business-1 I have 4 offers.
Now if I want to take into account the offers for some Profile, I have to make a join with my ternary relationship. But even if I do something as simple as the following
select distinct business_name
from Offer
INNER JOIN ternary ON Offer.id_Offer = ternary.id_Offer
GROUP BY business_name
WHERE business_name = 'business-1'
No matter what I put on the group by, or if I write distinct or not, I do not get what I want. The reality is that for business-1, I have 4 offers. Right now in the ternary only appear two. So it should return 2 unique offers for this name with no filtering by profile.
But instead I get 8 offers, because that is how many times it appears in the ternary, the id_Offer's that match.
How should this be done? If I need no filters I can simply look at Offers table alone. But what if I need to filter by id_skill or id_Profile AND want to return the business_name?
I have seen solutions such as this but I can not make them work, I do not understand what the ? is, how is it called to learn more about it, if MariaDB works the same in this sense, I could not find information about it because I do not know how that operation is called. When I try to build that query for my data I get:
ERROR 1064 (42000): You have an error in your SQL syntax; check the manual that corresponds to your MariaDB server version for the right syntax to use near '? ORDER BY COUNT(*) DESC' at line 1
But as I said, it is kind of hard to look for '?' as an... Operator? Function?

Answer 1

There are two basic solutions.
SELECT
o.business_name,
COUNT(DISTINCT o.id_offer) AS unique_offers
FROM
Offer AS o
INNER JOIN
ternary AS t
ON t.id_Offer = o.id_Offer
WHERE
o.business_name = 'business-1'
AND t.id_profile IN (1, 2, 3, 5)
GROUP BY
o.business_name
That's the simplest to write and think about. But, it can also be quite intensive because you're still joining each row in offer to 4 rows in ternary - Creating 8 rows to aggregate and process through DISTINCT.
The "better" (in my opinion) route is to filter then aggregate the ternary table in a sub-query.
SELECT
o.business_name,
COUNT(*) AS unique_offers
FROM
Offer AS o
INNER JOIN
(
SELECT id_Offer
FROM ternary
WHERE id_profile IN (1, 2, 3, 5)
GROUP BY id_Offer
)
AS t
ON t.id_Offer = o.id_Offer
WHERE
o.business_name = 'business-1'
GROUP BY
o.business_name
This ensures the t only ever has one row for any given offer. This in turn means that each row in offer only ever joins to one row in t; no duplication. That in turn means there is no need to use COUNT(DISTINCT) and relieves some overhead (By moving it to the inner query's GROUP BY).

Answer 2

Are you saying that you want to see offers for a particular business, but you want to limit these according to certain profiles or skills?
We limit query results in the WHERE clause. If we want to look up data in another table, we use IN or EXISTS. For instance:
select *
from offer
where business_name = 'business-1'
and id_offer in
(
select id_offer
from ternary
where id_profile = 1
and id_skill = 2
);