Find an exact duplicate by join table field - mysql

Question

I'm writing a query to find an exact duplicate of the order by its pruducts IDs.
The conditions to find a duplicate are:
1) Order has the same product count.
2) All product IDs are the same.
Tried something like this, but it didn't work:
SELECT
order.*,
count(same_products.id),
count(all_products.id)
FROM orders
LEFT JOIN products AS all_products ON all_products.order_id = orders.id
LEFT JOIN products AS same_products
ON same_products.order_id = orders.id AND same_products.id IN (30868, 30862)
GROUP BY orders.id
HAVING count(same_products.id) = 4 AND count(all_products.id = 4)

Answer 1

if you want count duplicated rows you should avoid the all columns (*) selector because if you have incremental values in your columns this don't let you find the duplicated .
SELECT
order.id
count(same_products.id),
count(all_products.id)
FROM orders
LEFT JOIN products AS all_products ON all_products.order_id = orders.id
LEFT JOIN products AS same_products
ON same_products.order_id = orders.id AND same_products.id IN (30868, 30862)
GROUP BY orders.id
HAVING count(same_products.id) >1 OR count(all_products.id )> 1
and for duplicated row you should check for count> 1 (for both the count)
and be careful with count(all_products.id =4) if you need to filter for this value you should add this to the on condition for the related table eg:
SELECT
order.id
count(same_products.id),
count(all_products.id)
FROM orders
LEFT JOIN products AS all_products ON all_products.order_id = orders.id and all_products.id =4
LEFT JOIN products AS same_products
ON same_products.order_id = orders.id AND same_products.id IN (30868, 30862)
GROUP BY orders.id
HAVING count(same_products.id) >1

Answer 2

I am not clear what you really mean by duplicate, I assume from your description that it's 2 orders with the same products. This seems a little simplistic for example given
MariaDB [sandbox]> select * from orders;
+------+---------------------+-------------+
| id | order_created | customer_id |
+------+---------------------+-------------+
| 1 | 2016-01-01 00:00:00 | 1 |
| 2 | 2016-02-01 00:00:00 | 1 |
| 3 | 2016-03-01 00:00:00 | 1 |
| 4 | 2016-01-01 00:00:00 | 2 |
| 5 | 2016-02-01 00:00:00 | 2 |
| 6 | 2016-01-01 00:00:00 | 3 |
| 10 | 2016-12-01 00:00:00 | 4 |
+------+---------------------+-------------+
7 rows in set (0.00 sec)
MariaDB [sandbox]> select * from order_details;
+----+---------+-----------+------+
| id | orderid | productid | qty |
+----+---------+-----------+------+
| 1 | 1 | 1213 | 10 |
| 2 | 1 | 9999 | 10 |
| 3 | 2 | 8888 | 10 |
| 4 | 3 | 1213 | 10 |
| 5 | 4 | 2222 | 10 |
| 6 | 5 | 9999 | 30 |
| 7 | 5 | 1213 | 30 |
| 8 | 6 | 9999 | 30 |
| 9 | 6 | 1213 | 30 |
+----+---------+-----------+------+
9 rows in set (0.00 sec)
select orders1.*,orders2.*,t.*
from
(select * from
(
select orderid o1orderid,group_concat(productid order by productid) o1grp,sum(qty) qty1
from order_details
group by orderid
) o1
join
(select orderid o2orderid,group_concat(productid order by productid) o2grp, sum(qty) qty2
from order_details
group by orderid
) o2
on o2grp = o1grp and qty2 = qty1 and o2orderid > o1orderid
) t
join orders orders1 on t.o1orderid = orders1.id
join orders orders2 on t.o2orderid = orders2.id
returns
+------+---------------------+-------------+------+---------------------+-------------+-----------+-----------+-------+-----------+-----------+-------+
| id | order_created | customer_id | id | order_created | customer_id | o1orderid | o1grp | qty1 | o2orderid | o2grp | qty2 |
+------+---------------------+-------------+------+---------------------+-------------+-----------+-----------+-------+-----------+-----------+-------+
| 5 | 2016-02-01 00:00:00 | 2 | 6 | 2016-01-01 00:00:00 | 3 | 5 | 1213,9999 | 30,30 | 6 | 1213,9999 | 30,30 |
+------+---------------------+-------------+------+---------------------+-------------+-----------+-----------+-------+-----------+-----------+-------+
1 row in set (0.03 sec)
But customer number is different.