[SPARK-20438][R] SparkR wrappers for split and repeat

R/pkg/NAMESPACE

@@ -300,6 +300,7 @@ exportMethods("%in%",
               "rank",
               "regexp_extract",
               "regexp_replace",
+              "repeat_string",
               "reverse",
               "rint",
               "rlike",
@@ -323,6 +324,7 @@ exportMethods("%in%",
               "sort_array",
               "soundex",
               "spark_partition_id",
+              "split_string",
               "stddev",
               "stddev_pop",
               "stddev_samp",

R/pkg/R/functions.R

@@ -3745,3 +3745,61 @@ setMethod("collect_set",
             jc <- callJStatic("org.apache.spark.sql.functions", "collect_set", x@jc)
             column(jc)
           })
+
+#' split_string
+#'
+#' Splits string on regular expression.
+#'
+#' Equivalent to \code{split} SQL function
+#'
+#' @param x Column to compute on
+#' @param pattern Java regular expression
+#'
+#' @rdname split_string
+#' @family string_funcs
+#' @aliases split_string,Column-method
+#' @export
+#' @examples \dontrun{
+#' df <- read.text("README.md")
+#'
+#' head(select(df, split_string(df$value, "\\s+")))
+#'
+#' # This is equivalent to the following SQL expression
+#' head(selectExpr(df, "split(value, '\\\\s+')"))
+#' }
+#' @note split_string 2.3.0
+setMethod("split_string",
+          signature(x = "Column", pattern = "character"),
+          function(x, pattern) {
+            jc <- callJStatic("org.apache.spark.sql.functions", "split", x@jc, pattern)
+            column(jc)
+          })
+
+#' repeat_string
+#'
+#' Repeats string n times.
+#'
+#' Equivalent to \code{repeat} SQL function
+#'
+#' @param x Column to compute on
+#' @param n Number of repetitions
+#'
+#' @rdname repeat_string
+#' @family string_funcs
+#' @aliases repeat_string,Column-method
+#' @export
+#' @examples \dontrun{
+#' df <- read.text("README.md")
+#'
+#' first(select(df, repeat_string(df$value, 3)))
+#'
+#' # This is equivalent to the following SQL expression
+#' first(selectExpr(df, "repeat(value, 3)"))
+#' }
+#' @note repeat_string 2.3.0
+setMethod("repeat_string",
+          signature(x = "Column", n = "numeric"),
+          function(x, n) {
+            jc <- callJStatic("org.apache.spark.sql.functions", "repeat", x@jc, numToInt(n))
+            column(jc)
+          })

R/pkg/R/generics.R

@@ -1192,6 +1192,10 @@ setGeneric("regexp_extract", function(x, pattern, idx) { standardGeneric("regexp
 setGeneric("regexp_replace",
            function(x, pattern, replacement) { standardGeneric("regexp_replace") })
 
+#' @rdname repeat_string
+#' @export
+setGeneric("repeat_string", function(x, n) { standardGeneric("repeat_string") })
+
 #' @rdname reverse
 #' @export
 setGeneric("reverse", function(x) { standardGeneric("reverse") })
@@ -1257,6 +1261,10 @@ setGeneric("skewness", function(x) { standardGeneric("skewness") })
 #' @export
 setGeneric("sort_array", function(x, asc = TRUE) { standardGeneric("sort_array") })
 
+#' @rdname split_string
+#' @export
+setGeneric("split_string", function(x, pattern) { standardGeneric("split_string") })
+
 #' @rdname soundex
 #' @export
 setGeneric("soundex", function(x) { standardGeneric("soundex") })

R/pkg/inst/tests/testthat/test_sparkSQL.R

@@ -1546,6 +1546,40 @@ test_that("string operators", {
   expect_equal(collect(select(df3, substring_index(df3$a, ".", 2)))[1, 1], "a.b")
   expect_equal(collect(select(df3, substring_index(df3$a, ".", -3)))[1, 1], "b.c.d")
   expect_equal(collect(select(df3, translate(df3$a, "bc", "12")))[1, 1], "a.1.2.d")
+
+  l4 <- list(list(a = "[email protected]   1\\b"))
+  df4 <- createDataFrame(l4)
+  expect_equal(
+    collect(select(df4, split_string(df4$a, "\\s+")))[1, 1],
+    list(list("[email protected]", "1\\b"))
+  )
+  expect_equal(
+    collect(select(df4, split_string(df4$a, "\\.")))[1, 1],
+    list(list("a", "b@c", "d   1\\b"))
+  )
+  expect_equal(
+    collect(select(df4, split_string(df4$a, "@")))[1, 1],
+    list(list("a.b", "c.d   1\\b"))
+  )
+  expect_equal(
+    collect(select(df4, split_string(df4$a, "\\\\")))[1, 1],
+    list(list("[email protected]   1", "b"))
+  )
+
+  l5 <- list(list(a = "abc"))
+  df5 <- createDataFrame(l5)
+  expect_equal(
+    collect(select(df5, repeat_string(df5$a, 1L)))[1, 1],
+    "abc"
+  )
+  expect_equal(
+    collect(select(df5, repeat_string(df5$a, 3)))[1, 1],
+    "abcabcabc"
+  )
+  expect_equal(
+    collect(select(df5, repeat_string(df5$a, -1)))[1, 1],
+    ""
+  )
 })
 
 test_that("date functions on a DataFrame", {