Merge remote-tracking branch 'upstream/master' into SPARK-5932

Author: Ilya Ganelin

R/pkg/DESCRIPTION

@@ -17,19 +17,19 @@ License: Apache License (== 2.0)
 Collate:
     'generics.R'
     'jobj.R'
-    'SQLTypes.R'
     'RDD.R'
     'pairRDD.R'
+    'schema.R'
     'column.R'
     'group.R'
     'DataFrame.R'
     'SQLContext.R'
+    'backend.R'
     'broadcast.R'
+    'client.R'
     'context.R'
     'deserialize.R'
     'serialize.R'
     'sparkR.R'
-    'backend.R'
-    'client.R'
     'utils.R'
     'zzz.R'

R/pkg/NAMESPACE

@@ -5,6 +5,7 @@ exportMethods(
               "aggregateByKey",
               "aggregateRDD",
               "cache",
+              "cartesian",
               "checkpoint",
               "coalesce",
               "cogroup",
@@ -28,6 +29,7 @@ exportMethods(
               "fullOuterJoin",
               "glom",
               "groupByKey",
+              "intersection",
               "join",
               "keyBy",
               "keys",
@@ -52,11 +54,14 @@ exportMethods(
               "reduceByKeyLocally",
               "repartition",
               "rightOuterJoin",
+              "sampleByKey",
               "sampleRDD",
               "saveAsTextFile",
               "saveAsObjectFile",
               "sortBy",
               "sortByKey",
+              "subtract",
+              "subtractByKey",
               "sumRDD",
               "take",
               "takeOrdered",
@@ -95,6 +100,7 @@ exportClasses("DataFrame")
 exportMethods("columns",
               "distinct",
               "dtypes",
+              "except",
               "explain",
               "filter",
               "groupBy",
@@ -118,7 +124,6 @@ exportMethods("columns",
               "show",
               "showDF",
               "sortDF",
-              "subtract",
               "toJSON",
               "toRDD",
               "unionAll",
@@ -178,5 +183,14 @@ export("cacheTable",
        "toDF",
        "uncacheTable")
 
-export("print.structType",
-       "print.structField")
+export("sparkRSQL.init",
+       "sparkRHive.init")
+
+export("structField",
+       "structField.jobj",
+       "structField.character",
+       "print.structField",
+       "structType",
+       "structType.jobj",
+       "structType.structField",
+       "print.structType")

R/pkg/R/DataFrame.R

@@ -17,7 +17,7 @@
 
 # DataFrame.R - DataFrame class and methods implemented in S4 OO classes
 
-#' @include jobj.R SQLTypes.R RDD.R pairRDD.R column.R group.R
+#' @include generics.R jobj.R schema.R RDD.R pairRDD.R column.R group.R
 NULL
 
 setOldClass("jobj")
@@ -1141,29 +1141,31 @@ setMethod("intersect",
             dataFrame(intersected)
           })
 
-#' Subtract
+#' except
 #'
 #' Return a new DataFrame containing rows in this DataFrame
 #' but not in another DataFrame. This is equivalent to `EXCEPT` in SQL.
 #'
 #' @param x A Spark DataFrame
 #' @param y A Spark DataFrame
-#' @return A DataFrame containing the result of the subtract operation.
-#' @rdname subtract
+#' @return A DataFrame containing the result of the except operation.
+#' @rdname except
 #' @export
 #' @examples
 #'\dontrun{
 #' sc <- sparkR.init()
 #' sqlCtx <- sparkRSQL.init(sc)
 #' df1 <- jsonFile(sqlCtx, path)
 #' df2 <- jsonFile(sqlCtx, path2)
-#' subtractDF <- subtract(df, df2)
+#' exceptDF <- except(df, df2)
 #' }
-setMethod("subtract",
+#' @rdname except
+#' @export
+setMethod("except",
           signature(x = "DataFrame", y = "DataFrame"),
           function(x, y) {
-            subtracted <- callJMethod(x@sdf, "except", y@sdf)
-            dataFrame(subtracted)
+            excepted <- callJMethod(x@sdf, "except", y@sdf)
+            dataFrame(excepted)
           })
 
 #' Save the contents of the DataFrame to a data source