spark-core-groupByKey-01-order-count-by-date.scala

val orders = sc.textFile("/public/retail_db/orders")

val ordersMap = orders.
  map(o => (o.split(",")(1), 1))
val ordersGroupByDate = ordersMap.
  groupByKey
val orderCountByDate = ordersGroupByDate.
  map(o => (o._1, o._2.size))

orderCountByDate.
  take(10).
  foreach(println)

spark-core-groupByKey-02-order-sort-by-customer-and-date-desc.scala

val orders = sc.textFile("/public/retail_db/orders")

val ordersMap = orders.
  map(o => (o.split(",")(2).toInt, o))
val ordersGroupByCustomerId = ordersMap.
  groupByKey
val ordersByCustomerByDateDesc = ordersGroupByCustomerId.
  flatMap(o => {
    val ordersPerCustomer = o._2
    ordersPerCustomer.
      toList.
      sortBy(k => k.split(",")(1))(Ordering.String.reverse)
  })