How to join 2 tables without relationship in SQL - sql

Question

I am struggling to combine three tables with outer join, but cant get it right
I am using PostgreSQL database..
My tables look like this:
My Orders table have the foreign key of Customers and Years (this two dont have any relation)
Some year and Customer are not presented in my Orders table
I am trying to make an sql query so that each customer have all year. I can get this result by cross join customer and year, but this approach doesn't work when i continue to join this result with other tables. (My Orders table have other foreign keys also that i have to join also) So can i get this result by using outer join instead?
I have tried:
Select* From Orders
RIGHT JOIN Customers on Orders.customer_id = customer.id
That give me all the customer in Orders table (even them who doesent have a order) then i would like to do the same with all years also so every customer have one row for each year(2015-2020) Have tryed to do another right join with years table but it doesent work.. Anyone know how to fix this?
p.s the names of the tables is not real,i just used this names to make it easier to understand!

Answer 1

If you want all customers to have a year and their respective orders if they have them per year.
SELECT CUS.ID AS CustomerID, CUS.Name AS CustomerName, YEA.year AS Year, ORD.ID AS OrderID
FROM Customers AS CUS
CROSS JOIN Years AS YEA
LEFT JOIN Orders AS ORD
ON CUS.ID = ORD.Customer_ID
AND ORD.YEAR_ID = YEA.ID;
This will give you and their orders as seperate rows.
If you want the number of orders use this instead:
SELECT CUS.ID AS CustomerID, CUS.Name AS CustomerName, YEA.year AS Year, Count(ORD.ID) AS NrOfOrders
FROM Customers AS CUS
CROSS JOIN Years AS YEA
LEFT JOIN Orders AS ORD
ON CUS.ID = ORD.Customer_ID
AND ORD.YEAR_ID = YEA.ID
GROUP BY CUS.ID, CUS.Name, YEA.Year;
Which will show the number of orders and give each customer a row per year.
Try it out/see it in action here, I added some data to show the multiple orders.

Answer 2

A CROSS JOIN of customers and years, then a LEFT JOIN of that with orders seems to do exactly what you describe in your post. Or am I getting something wrong?
And don't put your data into an image next time. Just type it /paste it into your post, so we can copy-paste it into our examples ... I had to re-type them...
And "name" and "year" are reserved words. I avoided them in my example.
\pset null NULL
WITH
-- your input ...
customers ( id, nm) AS (
SELECT 1,'alpha'
UNION ALL SELECT 2,'bravo'
UNION ALL SELECT 3,'charlie'
UNION ALL SELECT 4,'delta'
UNION ALL SELECT 5,'echo'
UNION ALL SELECT 6,'foxtrot'
)
,
years(id,yr) AS (
SELECT 1,2015
UNION ALL SELECT 2,2016
UNION ALL SELECT 3,2017
UNION ALL SELECT 4,2018
UNION ALL SELECT 5,2019
UNION ALL SELECT 6,2020
)
,
orders(id,cust_id,yr_id) AS (
SELECT 1,1,1
UNION ALL SELECT 2,2,3
UNION ALL SELECT 3,4,5
UNION ALL SELECT 4,5,6
)
-- end of your input ...
SELECT
cust.nm
, years.yr
, ord.id
FROM customers cust
CROSS JOIN years
LEFT JOIN orders ord ON ord.yr_id=years.id
ORDER BY 1,2
;
-- out nm | yr | id
-- out ---------+------+------
-- out alpha | 2015 | 1
-- out alpha | 2016 | NULL
-- out alpha | 2017 | 2
-- out alpha | 2018 | NULL
-- out alpha | 2019 | 3
-- out alpha | 2020 | 4
-- out bravo | 2015 | 1
-- out bravo | 2016 | NULL
-- out bravo | 2017 | 2
-- out bravo | 2018 | NULL
-- out bravo | 2019 | 3
-- out bravo | 2020 | 4
-- out charlie | 2015 | 1
-- out charlie | 2016 | NULL
-- out charlie | 2017 | 2
-- out charlie | 2018 | NULL
-- out charlie | 2019 | 3
-- out charlie | 2020 | 4
-- out delta | 2015 | 1
-- out delta | 2016 | NULL
-- out delta | 2017 | 2
-- out delta | 2018 | NULL
-- out delta | 2019 | 3
-- out delta | 2020 | 4
-- out echo | 2015 | 1
-- out echo | 2016 | NULL
-- out echo | 2017 | 2
-- out echo | 2018 | NULL
-- out echo | 2019 | 3
-- out echo | 2020 | 4
-- out foxtrot | 2015 | 1
-- out foxtrot | 2016 | NULL
-- out foxtrot | 2017 | 2
-- out foxtrot | 2018 | NULL
-- out foxtrot | 2019 | 3
-- out foxtrot | 2020 | 4