JPA Query with GROUP BY, HAVING and COUNT - jpa

Question

So the query below is probably not the most efficient, buy still, I am wondering why it is returning no result, even though the SQL counterpart does. There is no error, I am just getting no result. Is it maybe not the correct equivalent for the query I wrote in MySQL?
This is the JPA JPQL.
Query query = em.createQuery("SELECT sub FROM Subscription sub WHERE "
+ "sub.isSuspended = 0 AND "
+ "(SELECT i FROM Invoice i WHERE i.dateDue < CURRENT_DATE AND i.datePaid IS NULL "
+ "GROUP BY i HAVING COUNT(i.idInvoice) > 2) MEMBER OF sub.invoices");
And this is the SQL from MySQL.
SELECT * from subscription
WHERE subscription.is_suspended = 0 AND id_subscription IN
(SELECT id_subscription FROM invoice
WHERE date_due < CURDATE() AND date_paid IS NULL
GROUP BY id_subscription
HAVING COUNT(*) > 2)

Answer 1

The two queries are not the same. To use the actual query use the NativeQuery createNativeQuery() instead of Query.
In your case the JPA version seems to have syntax errors.
After the AND you are missing the IN operator.
In the nested query you are selecting i instead of something like i.idInvoice
The JPA query should look like
SELECT sub FROM Subscription sub
WHERE sub.isSuspended = 0
AND sub.idSubscription IN
(SELECT i.idInvoice
FROM Invoice i
WHERE i.dateDue < CURRENT_DATE AND i.datePaid IS NULL
GROUP BY i.idInvoice
HAVING COUNT(i.idInvoice) > 2);