PySpark - deal with column renaming and data type changes

 Data Column name mapping

if need_mapping == "true":

  df_transformed = df_raw.select(csv_cols_list)
  for csv_col,rpt_col in zip(csv_cols_list, report_cols_list):
      df_transformed = df_transformed.withColumnRenamed(csv_col,rpt_col)
else:
  df_transformed = df_raw

Data type changes:

e.g. string value like ($3,400) change to -3400

for item in numeric_cols_list:
    df = df.withColumn(item, when(instr(col(item), "(") > 0,
             concat(F.lit('-'),F.regexp_replace(item, "[(,\$#),]",'')))
            .otherwise(F.regexp_replace(item, "[(,\$#),]", '')).cast("double"))