how spark identified duplicate column name

Question

I have a simple question, about spark behavior

//scala
> df.printSchema
root
 |-- ID0: string (nullable = true)
 |-- id1: string (nullable = true)
 |-- Name: s