with_column.py

#   Read the source tables in Parquet format
sales_table = spark.read.parquet("./data/sales_parquet")

'''
SELECT order_id,
       product_id,
       seller_id,
       date,
       num_pieces_sold,
       bill_raw_text,
       num_pieces_sold % 2 AS num_pieces_sold_is_even
FROM sales_table a
'''
sales_table_execution_plan = sales_table. \
    withColumn("num_pieces_sold_is_even", col("num_pieces_sold")%2)

sales_table_execution_plan.show()