Pyspark quando várias condições

df5.withColumn("new_column", when(col("code") == "a" | col("code") == "d", "A")
      .when(col("code") == "b" & col("amt") == "4", "B")
      .otherwise("A1")).show()
Unsightly Unicorn