Question

I don’t mean to select duplicate records only once.
Instead I wont to select those records from table, which are not duplicate. Records present in table only once.

For example:
in table cat_prod there are columns id_category and id_product.
One product can be in multiple categories.
How can I select only those product, which are only in one category?

Answer 1

SELECT id_product
FROM cat_prod
GROUP BY id_product
HAVING count(distinct id_category) = 1

You can group by the product id and select only those that have COUNT(*) = 1 in the HAVING clause.