Let's say you have a table oldtable with the following fields:
create table oldtable (product_id int, product_name varchar(100), product_size varchar(100));
Now you realize you have duplicate product_ids that may or may not have the same values for other fields. And you want to de-duplicate your table by picking only one of the rows.
You can do this is mysql very simply:
create table newtable select * from oldtable where 1 group by product_id
This will give you unique product_ids with the product_name and product_size associated with the first occurrence of the product_id