spark udf accepts and handles null values.

Difficulty

It is convenient to deal with null of string column in spark, but it is very inconvenient to deal with int double and so on.

You may say that I can use Option[Int], but no, udf does not support this.

You will say that I accept it with Any, and then judge whether it is null or Int, but it still doesn't work.

Here's my solution:

-------------------------

// _ooOoo_//o8888888o// 88" . "88// (| -_- |)// O\ = /O// ____/`---'\____// . ' \\| |// `.// / \\||| : |||// \// / _||||| -:- |||||- \// | | \\\ - /// | |// | \_| ''\---/'' | |// \ .-\__ `-` ___/-. /// ___`. .' /--.--\ `. . __//