Hello!
Thank you for taking time out of your busy schedule to read my email,My English is no good. The company data is ciphertext and needs to be grouped by calling the interface. Similar to two ciphertext data passed to the interface, the interface returns true or false to determine whether the two data are in the same group. But Spark doesn't seem to have an apI to change the grouping rules; If the ciphertext is not grouped by the interface, the ciphertext is stored as a string. Spark groups the ciphertext according to the comparison rules of strings, resulting in incorrect grouping So I need to be able to rewrite the grouping rules and call the company's interface instead. Thanks.
