Merge pull request apache#109 from hlin09/hlin09

[SPARKR-15] Rename the filter function to filterRDD to follow the API consistency.

Author: concretevitamin

pkg/NAMESPACE

@@ -13,7 +13,7 @@ exportMethods(
               "countByValue",
               "distinct",
               "Filter",
-              "filter",
+              "filterRDD",
               "flatMap",
               "flatMapValues",
               "groupByKey",

pkg/R/RDD.R

@@ -600,39 +600,36 @@ setMethod("mapPartitionsWithIndex",
 #' a predicate (i.e. returning TRUE in a given logical function). 
 #' The same as `filter()' in Spark.
 #'
-#' @param f A unary predicate function.
-#' @param x The RDD to be filtered.
-#' @rdname Filter
+#' @param rdd The RDD to be filtered.
+#' @param filterFunc A unary predicate function.
+#' @rdname filterRDD
 #' @export
 #' @examples
 #'\dontrun{
 #' sc <- sparkR.init()
 #' rdd <- parallelize(sc, 1:10)
-#' unlist(collect(Filter(function (x) { x < 3 }, rdd))) # c(1, 2)
+#' unlist(collect(filterRDD(rdd, function (x) { x < 3 }))) # c(1, 2)
 #'}
-#setGeneric("Filter", function(f, x) { standardGeneric("Filter") })
-
-#' @rdname Filter
-#' @aliases Filter,function,RDD-method filter,function,RDD-method
-setMethod("Filter",
-          signature(f = "function", x = "RDD"),
-          function(f, x) {
+setGeneric("filterRDD", 
+           function(rdd, filterFunc) { standardGeneric("filterRDD") })
+
+#' @rdname filterRDD
+#' @aliases filterRDD,RDD,function-method
+setMethod("filterRDD",
+          signature(rdd = "RDD", filterFunc = "function"),
+          function(rdd, filterFunc) {
             filter.func <- function(part) {
-              Filter(f, part)
+              Filter(filterFunc, part)
             }
-            lapplyPartition(x, filter.func)
+            lapplyPartition(rdd, filter.func)
           })
 
-#' @rdname Filter
-#' @export
-setGeneric("filter", function(f, x) { standardGeneric("filter") })
-
-#' @rdname Filter
-#' @aliases filter,function,RDD-method
-setMethod("filter",
+#' @rdname filterRDD
+#' @aliases Filter,function,RDD-method
+setMethod("Filter",
           signature(f = "function", x = "RDD"),
           function(f, x) {
-            Filter(f, x)
+            filterRDD(x, f)
           })
 
 #' Reduce across elements of an RDD.

pkg/inst/tests/test_rdd.R

@@ -55,8 +55,8 @@ test_that("flatMap() on RDDs", {
   expect_equal(actual, rep(intPairs, each=2))
 })
 
-test_that("Filter on RDD", {
-  filtered.rdd <- Filter(function(x) { x %% 2 == 0 }, rdd)
+test_that("filterRDD on RDD", {
+  filtered.rdd <- filterRDD(rdd, function(x) { x %% 2 == 0 })
   actual <- collect(filtered.rdd)
   expect_equal(actual, list(2, 4, 6, 8, 10))
 
@@ -65,7 +65,7 @@ test_that("Filter on RDD", {
   expect_equal(actual, list(list(1L, -1)))
 
   # Filter out all elements.
-  filtered.rdd <- Filter(function(x) { x > 10 }, rdd)
+  filtered.rdd <- filterRDD(rdd, function(x) { x > 10 })
   actual <- collect(filtered.rdd)
   expect_equal(actual, list())
 })

pkg/man/Filter.Rd renamed to pkg/man/filterRDD.Rd

@@ -1,23 +1,23 @@
 % Generated by roxygen2 (4.0.2): do not edit by hand
 \docType{methods}
-\name{Filter,function,RDD-method}
+\name{filterRDD}
 \alias{Filter,function,RDD-method}
-\alias{filter}
-\alias{filter,function,RDD-method}
+\alias{filterRDD}
+\alias{filterRDD,RDD,function-method}
 \title{This function returns a new RDD containing only the elements that satisfy
 a predicate (i.e. returning TRUE in a given logical function).
 The same as `filter()' in Spark.}
 \usage{
-\S4method{Filter}{`function`,RDD}(f, x)
+filterRDD(rdd, filterFunc)
 
-filter(f, x)
+\S4method{filterRDD}{RDD,`function`}(rdd, filterFunc)
 
-\S4method{filter}{`function`,RDD}(f, x)
+\S4method{Filter}{`function`,RDD}(f, x)
 }
 \arguments{
-\item{f}{A unary predicate function.}
+\item{rdd}{The RDD to be filtered.}
 
-\item{x}{The RDD to be filtered.}
+\item{filterFunc}{A unary predicate function.}
 }
 \description{
 This function returns a new RDD containing only the elements that satisfy
@@ -28,7 +28,7 @@ The same as `filter()' in Spark.
 \dontrun{
 sc <- sparkR.init()
 rdd <- parallelize(sc, 1:10)
-unlist(collect(Filter(function (x) { x < 3 }, rdd))) # c(1, 2)
+unlist(collect(filterRDD(rdd, function (x) { x < 3 }))) # c(1, 2)
 }
 }