I have been dealing with an issue related to writing out a Parquet file in Spark, when the input file is Parquet as well and contains some invalid column names.
Unfor
