Merge pull request apache#144 from shivaram/fix-rd-files

[SPARKR-176] Fix SparkR Rd files

Author: concretevitamin

pkg/NAMESPACE

@@ -1,6 +1,5 @@
 #exportPattern("^[[:alpha:]]+")
 exportClasses("RDD")
-exportClasses("PipelinedRDD")
 exportClasses("Broadcast")
 exportMethods(
               "cache",
@@ -67,3 +66,4 @@ export(
 export("sparkR.init")
 export("print.jobj")
 useDynLib(SparkR, stringHashCode)
+importFrom(methods, setGeneric, setMethod, setOldClass)

pkg/R/RDD.R

@@ -8,8 +8,8 @@ setOldClass("jobj")
 #' @rdname RDD
 #' @seealso parallelize, textFile
 #'
-#' @param env An R environment that stores bookkeeping states of the RDD
-#' @param jrdd Java object reference to the backing JavaRDD
+#' @slot env An R environment that stores bookkeeping states of the RDD
+#' @slot jrdd Java object reference to the backing JavaRDD
 #' @export
 setClass("RDD",
          slots = list(env = "environment",
@@ -81,13 +81,16 @@ setMethod("initialize", "PipelinedRDD", function(.Object, prev, func, jrdd_val)
 
 #' @rdname RDD
 #' @export
+#'
+#' @param jrdd Java object reference to the backing JavaRDD
+#' @param serialized TRUE if the RDD stores data serialized in R
+#' @param isCached TRUE if the RDD is cached
+#' @param isCheckpointed TRUE if the RDD has been checkpointed
 RDD <- function(jrdd, serialized = TRUE, isCached = FALSE,
                 isCheckpointed = FALSE) {
   new("RDD", jrdd, serialized, isCached, isCheckpointed)
 }
 
-#' @rdname PipelinedRDD
-#' @export
 PipelinedRDD <- function(prev, func) {
   new("PipelinedRDD", prev, func, NULL)
 }
@@ -414,7 +417,7 @@ setMethod("lookup",
 
 #' Return the number of elements in the RDD.
 #'
-#' @param rdd The RDD to count
+#' @param x The RDD to count
 #' @return number of elements in the RDD.
 #' @rdname count
 #' @export
@@ -425,17 +428,17 @@ setMethod("lookup",
 #' count(rdd) # 10
 #' length(rdd) # Same as count
 #'}
-setGeneric("count", function(rdd) { standardGeneric("count") })
+setGeneric("count", function(x) { standardGeneric("count") })
 
 #' @rdname count
 #' @aliases count,RDD-method
 setMethod("count",
-          signature(rdd = "RDD"),
-          function(rdd) {
+          signature(x = "RDD"),
+          function(x) {
             countPartition <- function(part) {
               as.integer(length(part))
             }
-            valsRDD <- lapplyPartition(rdd, countPartition)
+            valsRDD <- lapplyPartition(x, countPartition)
             vals <- collect(valsRDD)
             sum(as.integer(vals))
           })
@@ -511,6 +514,7 @@ setMethod("countByKey",
 #' @param FUN the transformation to apply on each element
 #' @return a new RDD created by the transformation.
 #' @rdname lapply
+#' @aliases lapply
 #' @export
 #' @examples
 #'\dontrun{
@@ -666,8 +670,8 @@ setMethod("mapPartitionsWithIndex",
 #' a predicate (i.e. returning TRUE in a given logical function).
 #' The same as `filter()' in Spark.
 #'
-#' @param rdd The RDD to be filtered.
-#' @param filterFunc A unary predicate function.
+#' @param x The RDD to be filtered.
+#' @param f A unary predicate function.
 #' @rdname filterRDD
 #' @export
 #' @examples
@@ -677,21 +681,22 @@ setMethod("mapPartitionsWithIndex",
 #' unlist(collect(filterRDD(rdd, function (x) { x < 3 }))) # c(1, 2)
 #'}
 setGeneric("filterRDD", 
-           function(rdd, filterFunc) { standardGeneric("filterRDD") })
+           function(x, f) { standardGeneric("filterRDD") })
 
 #' @rdname filterRDD
 #' @aliases filterRDD,RDD,function-method
 setMethod("filterRDD",
-          signature(rdd = "RDD", filterFunc = "function"),
-          function(rdd, filterFunc) {
+          signature(x = "RDD", f = "function"),
+          function(x, f) {
             filter.func <- function(part) {
-              Filter(filterFunc, part)
+              Filter(f, part)
             }
-            lapplyPartition(rdd, filter.func)
+            lapplyPartition(x, filter.func)
           })
 
 #' @rdname filterRDD
-#' @aliases Filter,function,RDD-method
+#' @export
+#' @aliases Filter
 setMethod("Filter",
           signature(f = "function", x = "RDD"),
           function(f, x) {
@@ -802,9 +807,6 @@ setMethod("foreach",
 
 #' Applies a function to each partition in an RDD, and force evaluation.
 #'
-#' @param rdd The RDD to apply the function
-#' @param func The function to be applied to partitions.
-#' @return invisible NULL.
 #' @export
 #' @rdname foreach
 #' @examples
@@ -1652,7 +1654,8 @@ setMethod("join",
 #' sc <- sparkR.init()
 #' rdd1 <- parallelize(sc, list(list(1, 1), list(2, 4)))
 #' rdd2 <- parallelize(sc, list(list(1, 2), list(1, 3)))
-#' leftOuterJoin(rdd1, rdd2, 2L) # list(list(1, list(1, 2)), list(1, list(1, 3)), list(2, list(4, NULL)))
+#' leftOuterJoin(rdd1, rdd2, 2L)
+#' # list(list(1, list(1, 2)), list(1, list(1, 3)), list(2, list(4, NULL)))
 #'}
 setGeneric("leftOuterJoin", function(rdd1, rdd2, numPartitions) { standardGeneric("leftOuterJoin") })
 
@@ -1721,7 +1724,8 @@ setMethod("leftOuterJoin",
 #' sc <- sparkR.init()
 #' rdd1 <- parallelize(sc, list(list(1, 2), list(1, 3)))
 #' rdd2 <- parallelize(sc, list(list(1, 1), list(2, 4)))
-#' rightOuterJoin(rdd1, rdd2, 2L) # list(list(1, list(2, 1)), list(1, list(3, 1)), list(2, list(NULL, 4)))
+#' rightOuterJoin(rdd1, rdd2, 2L)
+#' # list(list(1, list(2, 1)), list(1, list(3, 1)), list(2, list(NULL, 4)))
 #'}
 setGeneric("rightOuterJoin", function(rdd1, rdd2, numPartitions) { standardGeneric("rightOuterJoin") })
 

pkg/R/jobj.R

@@ -33,10 +33,17 @@ jobj <- function(objId) {
   obj
 }
 
-print.jobj <- function(jobj) {
-  cls <- callJMethod(jobj, "getClass")
+#' Print a JVM object reference.
+#'
+#' This function prints the type and id for an object stored
+#' in the SparkR JVM backend.
+#'
+#' @param x The JVM object reference
+#' @param ... further arguments passed to or from other methods
+print.jobj <- function(x, ...) {
+  cls <- callJMethod(x, "getClass")
   name <- callJMethod(cls, "getName")
-  cat("Java ref type", name, "id", jobj$id, "\n", sep = " ")
+  cat("Java ref type", name, "id", x$id, "\n", sep = " ")
 }
 
 cleanup.jobj <- function(jobj) {

pkg/R/serialize.R

@@ -51,7 +51,7 @@ writeBoolean <- function(con, value) {
 }
 
 writeRawSerialize <- function(outputCon, batch) {
-  outputSer <- serialize(batch, ascii = FALSE, conn = NULL)
+  outputSer <- serialize(batch, ascii = FALSE, connection = NULL)
   writeRaw(outputCon, outputSer)
 }
 

pkg/R/sparkR.R

@@ -59,6 +59,8 @@ sparkR.stop <- function(env) {
 #' @param sparkEnvir Named list of environment variables to set on worker nodes.
 #' @param sparkExecutorEnv Named list of environment variables to be used when launching executors.
 #' @param sparkJars Character string vector of jar files to pass to the worker nodes.
+#' @param sparkRLibDir The path where R is installed on the worker nodes.
+#' @param sparkRBackendPort The port to use for SparkR JVM Backend.
 #' @export
 #' @examples
 #'\dontrun{
@@ -67,7 +69,7 @@ sparkR.stop <- function(env) {
 #'                  list(spark.executor.memory="1g"))
 #' sc <- sparkR.init("yarn-client", "SparkR", "/home/spark",
 #'                  list(spark.executor.memory="1g"),
-#'                  list(LD_LIBRARY_PATH="/directory of Java VM Library Files (libjvm.so) on worker nodes/"),
+#'                  list(LD_LIBRARY_PATH="/directory of JVM libraries (libjvm.so) on workers/"),
 #'                  c("jarfile1.jar","jarfile2.jar"))
 #'}
 

pkg/R/sparkRBackend.R

@@ -4,8 +4,8 @@
 # Returns TRUE if object is an instance of given class
 isInstanceOf <- function(jobj, className) {
   stopifnot(class(jobj) == "jobj")
-  cls <- SparkR:::callJStatic("java.lang.Class", "forName", className)
-  SparkR:::callJMethod(cls, "isInstance", jobj)
+  cls <- callJStatic("java.lang.Class", "forName", className)
+  callJMethod(cls, "isInstance", jobj)
 }
 
 # Call a Java method named methodName on the object

pkg/man/RDD.Rd

@@ -1,21 +1,32 @@
+% Generated by roxygen2 (4.0.2): do not edit by hand
 \docType{class}
 \name{RDD-class}
 \alias{RDD}
 \alias{RDD-class}
 \title{S4 class that represents an RDD}
 \usage{
-RDD(jrdd, serialized = TRUE)
+RDD(jrdd, serialized = TRUE, isCached = FALSE, isCheckpointed = FALSE)
 }
 \arguments{
-  \item{jrdd}{Java object reference to the backing JavaRDD}
+\item{jrdd}{Java object reference to the backing JavaRDD}
 
-  \item{serialized}{TRUE if the JavaRDD contains serialized
-  R objects}
+\item{serialized}{TRUE if the RDD stores data serialized in R}
+
+\item{isCached}{TRUE if the RDD is cached}
+
+\item{isCheckpointed}{TRUE if the RDD has been checkpointed}
 }
 \description{
-RDD can be created using functions like \code{parallelize},
-\code{textFile} etc.
+RDD can be created using functions like
+             \code{parallelize}, \code{textFile} etc.
 }
+\section{Slots}{
+
+\describe{
+\item{\code{env}}{An R environment that stores bookkeeping states of the RDD}
+
+\item{\code{jrdd}}{Java object reference to the backing JavaRDD}
+}}
 \seealso{
 parallelize, textFile
 }

pkg/man/broadcast-class.Rd

@@ -1,3 +1,4 @@
+% Generated by roxygen2 (4.0.2): do not edit by hand
 \docType{class}
 \name{Broadcast-class}
 \alias{Broadcast}
@@ -7,18 +8,17 @@
 Broadcast(id, value, jBroadcastRef, objName)
 }
 \arguments{
-  \item{id}{Id of the backing Spark broadcast variable}
+\item{id}{Id of the backing Spark broadcast variable}
 
-  \item{value}{Value of the broadcast variable}
+\item{value}{Value of the broadcast variable}
 
-  \item{jBroadcastRef}{reference to the backing Java
-  broadcast object}
+\item{jBroadcastRef}{reference to the backing Java broadcast object}
 
-  \item{objName}{name of broadcasted object}
+\item{objName}{name of broadcasted object}
 }
 \description{
 Broadcast variables can be created using the broadcast
-function from a \code{SparkContext}.
+             function from a \code{SparkContext}.
 }
 \seealso{
 broadcast

pkg/man/broadcast-internal.Rd

@@ -1,18 +1,18 @@
+% Generated by roxygen2 (4.0.2): do not edit by hand
 \name{setBroadcastValue}
 \alias{setBroadcastValue}
 \title{Internal function to set values of a broadcast variable.}
 \usage{
 setBroadcastValue(bcastId, value)
 }
 \arguments{
-  \item{bcastId}{The id of broadcast variable to set}
+\item{bcastId}{The id of broadcast variable to set}
 
-  \item{value}{The value to be set}
+\item{value}{The value to be set}
 }
 \description{
-This function is used internally by Spark to set the value
-of a broadcast variable on workers. Not intended for use
-outside the package.
+This function is used internally by Spark to set the value of a broadcast
+variable on workers. Not intended for use outside the package.
 }
 \seealso{
 broadcast, value

pkg/man/broadcast.Rd

@@ -1,3 +1,4 @@
+% Generated by roxygen2 (4.0.2): do not edit by hand
 \docType{methods}
 \name{value}
 \alias{broadcast}
@@ -12,19 +13,18 @@ value(bcast)
 broadcast(sc, object)
 }
 \arguments{
-  \item{bcast}{The broadcast variable to get}
+\item{bcast}{The broadcast variable to get}
 
-  \item{sc}{Spark Context to use}
+\item{sc}{Spark Context to use}
 
-  \item{object}{Object to be broadcast}
+\item{object}{Object to be broadcast}
 }
 \description{
-\code{value} can be used to get the value of a broadcast
-variable inside a distributed function.
+\code{value} can be used to get the value of a broadcast variable inside
+a distributed function.
 
-Broadcast a read-only variable to the cluster, returning a
-\code{Broadcast} object for reading it in distributed
-functions.
+Broadcast a read-only variable to the cluster, returning a \code{Broadcast}
+object for reading it in distributed functions.
 }
 \examples{
 \dontrun{