windowPartitionBy — windowPartitionBy • SparkR

Creates a WindowSpec with the partitioning defined.

Usage

windowPartitionBy(col, ...)

# S4 method for class 'character'
windowPartitionBy(col, ...)

# S4 method for class 'Column'
windowPartitionBy(col, ...)

Arguments

col: A column name or Column by which rows are partitioned to windows.
...: Optional column names or Columns in addition to col, by which rows are partitioned to windows.

Note

windowPartitionBy(character) since 2.0.0

windowPartitionBy(Column) since 2.0.0

Examples

if (FALSE) { # \dontrun{
  ws <- orderBy(windowPartitionBy("key1", "key2"), "key3")
  df1 <- select(df, over(lead("value", 1), ws))

  ws <- orderBy(windowPartitionBy(df$key1, df$key2), df$key3)
  df1 <- select(df, over(lead("value", 1), ws))
} # }