spark.version

import java.sql.Date
import org.apache.spark.sql.expressions.Window
import org.apache.spark.sql.functions._

val purchasesDF = Seq(
    (Date.valueOf("2019-01-01"), "01"),
    (Date.valueOf("2019-05-10"), "01"),
    (Date.valueOf("2019-03-05"), "02"),
    (Date.valueOf("2019-02-20"), "03"),
    (Date.valueOf("2019-01-20"), "02")
).toDF("purchase_date", "device_id")

val devicesDF = Seq(
    ("01", "notebook", 600.00),
    ("02", "small phone", 100.00),
    ("03", "camera",150.00),
    ("04", "large phone", 700.00)
).toDF("device_id", "device_name", "price")

%%dataframe
purchasesDF.join(devicesDF, col("device_id") === col("device_id"), "inner")

%%dataframe
%%scan
purchasesDF.as("purchases")
  .join(devicesDF.as("devices"), col("purchases.device_id") === col("devices.device_id"))

purchasesDF.as("purchases")
  .join(devicesDF.as("devices"), col("purchases.device_id") === col("devices.device_id"))
  .select("xxxx")

%%dataframe
%%scan
purchasesDF.as("purchases")
  .join(devicesDF.as("devices"), col("purchases.device_id") === col("devices.device_id"))
  .drop("device_id")