SQL - Rows not exist as exist

I have 3 tables. One with a list of users, one with items and one with relation between the users and the items. User can have an item (stat = 0/1) or can have a duplicate (dup = 0/1). If you click on the item's or the duplicate's button, it runs a query that adds or updates the row.

It looks like this:

| user | item_id | stat | dup |
| ---- | ------- | ---- | --- |
| 17   |   1     |  1   |  1  |
| 5    |   2     |  1   |  0  |
| 8    |   1     |  0   |  1  |
| 9    |   4     |  1   |  0  |

I compare these data who have an item what the other does not have and vica versa with this query:

SELECT t2.user_id,
       GROUP_CONCAT(CASE WHEN t1.dup=1 AND t2.stat=0 THEN item_id END) `My Item List`,
       GROUP_CONCAT(CASE WHEN t2.dup=1 AND t1.stat=0 THEN item_id END) `Item List`
FROM table3 t1
LEFT JOIN table3 t2 USING (item_id)
WHERE t1.user_id = @current_user
  AND t2.user_id <> @current_user
GROUP BY t2.user_id

My problem is I can only compare the rows if the user already clicked the button of the item at least once, so the row generated, but I need the stat=0 rows as well even if they are not existing.

After the code runs the table should look like this, so I can see the relations even if the user doesnt have (stat=0) that item:

| user | item_id | stat | dup |
| ---- | ------- | ---- | --- |
| 17   |   1     |  1   |  1  |
| 5    |   2     |  0   |  0  |
| 5    |   6     |  1   |  0  |
| 8    |   1     |  0   |  1  |
| 9    |   4     |  1   |  0  |
| 9    |   6     |  0   |  0  |
| 17   |   13    |  0   |  0  |

How can I integrate this to the query above?

My tables:

Table1

CREATE TABLE table1(
    id NOT NULL AUTO_INCREMENT,
    user_name varchar(255),
);

Table 2

CREATE TABLE table2(
    id NOT NULL AUTO_INCREMENT,
    item_no varchar(255),
    item_name varchar(255),
    item_group varchar(255)
);

Table 3

CREATE TABLE table3 (
  id int NOT NULL AUTO_INCREMENT,
  user_id int NOT NULL,
  item_id int NOT NULL,
  stat tinyint NOT NULL,
  dup tinyint NOT NULL,
  PRIMARY KEY (id),
  FOREIGN KEY (user_id) REFERENCES table1(id),
  FOREIGN KEY (item_id) REFERENCES table2(id)
);

Thank you!

What if you just grab all unique combinations of user and item_id, and then only map the combinations present and leave the rest as null? We can replace the null with 0.

Something like,

SELECT 
  all_combos.user, 
  all_combos.item_id, 
  COALESCE(T.stat, 0) AS stat, 
  COALESCE(T.dup) AS dup 
FROM 
  (
    SELECT 
      user, 
      item_id 
    FROM 
      (SELECT DISTINCT user FROM table3) u, 
      (SELECT DISTINCT item_id FROM table3) i
  ) AS all_combos 
  LEFT OUTER JOIN table3 T ON T.user = all_combos.user 
  AND T.item_id = all_combos.item_id

If you want to create combinations of users and items that might not have ever happened in table3, you can combine them by modifying the subqueries for u and i

SELECT 
  all_combos.user, 
  all_combos.item_id, 
  COALESCE(T.stat, 0) AS stat, 
  COALESCE(T.dup) AS dup 
FROM 
  (
    SELECT 
      user, 
      item_id 
    FROM 
      (SELECT DISTINCT id as user FROM table1) u, 
      (SELECT DISTINCT id as item_id FROM table2) i
  ) AS all_combos 
  LEFT OUTER JOIN table3 T ON T.user = all_combos.user 
  AND T.item_id = all_combos.item_id