spark sql window function lag

问题:

I am looking at the window slide function for a Spark DataFrame in Spark SQL, Scala.

I have a dataframe with columns Col1,Col1,Col1,date.

Col1    Col2    Col3    date     volume new_col
                        201601  100.5   
                        201602  120.6   100.5
                        201603  450.2   120.6
                        201604  200.7   450.2
                        201605  121.4   200.7`

Now I want to add a new column with name(new_col) with one row slided down, as shown above.

I tried below option to use the window function.

val windSldBrdrxNrx_df = df.withColumn("Prev_brand_rx", lag("Prev_brand_rx",1))

Can anyone please help me to how to do this.

回答1:

You are doing correctly all you missed is over(window expression) on lag

val df = sc.parallelize(Seq((201601, 100.5),
  (201602, 120.6),
  (201603, 450.2),
  (201604, 200.7),
  (201605, 121.4))).toDF("date", "volume")

val w = org.apache.spark.sql.expressions.Window.orderBy("date")  

import org.apache.spark.sql.functions.lag

val leadDf = df.withColumn("new_col", lag("volume", 1, 0).over(w))

leadDf.show()

+------+------+-------+
|  date|volume|new_col|
+------+------+-------+
|201601| 100.5|    0.0|
|201602| 120.6|  100.5|
|201603| 450.2|  120.6|
|201604| 200.7|  450.2|
|201605| 121.4|  200.7|
+------+------+-------+

This code was run on Spark shell 2.0.2

回答2:

You can import below two packages, which will resolve the issue of lag dependencies.

import org.apache.spark.sql.functions.{lead, lag}
import org.apache.spark.sql.expressions.Window