[SPARK-12436][SQL] Now InferSchema should return NullType and ArrayType(NullType)

.gitignore

@@ -60,7 +60,6 @@ dev/create-release/*final
 spark-*-bin-*.tgz
 unit-tests.log
 /lib/
-ec2/lib/
 rat-results.txt
 scalastyle.txt
 scalastyle-output.xml

.rat-excludes

@@ -84,3 +84,5 @@ gen-java.*
 org.apache.spark.sql.sources.DataSourceRegister
 org.apache.spark.scheduler.SparkHistoryListenerFactory
 .*parquet
+LZ4BlockInputStream.java
+spark-deps-.*

R/pkg/.lintr

@@ -1,2 +1,2 @@
-linters: with_defaults(line_length_linter(100), camel_case_linter = NULL, open_curly_linter(allow_single_line = TRUE), closed_curly_linter(allow_single_line = TRUE), commented_code_linter = NULL)
+linters: with_defaults(line_length_linter(100), camel_case_linter = NULL, open_curly_linter(allow_single_line = TRUE), closed_curly_linter(allow_single_line = TRUE))
 exclusions: list("inst/profile/general.R" = 1, "inst/profile/shell.R")

R/pkg/DESCRIPTION

@@ -1,7 +1,7 @@
 Package: SparkR
 Type: Package
 Title: R frontend for Spark
-Version: 1.6.0
+Version: 2.0.0
 Date: 2013-09-09
 Author: The Apache Software Foundation
 Maintainer: Shivaram Venkataraman <shivaram@cs.berkeley.edu>

R/pkg/NAMESPACE

@@ -94,7 +94,8 @@ exportMethods("arrange",
               "withColumnRenamed",
               "write.df",
               "write.json",
-              "write.parquet")
+              "write.parquet",
+              "write.text")
 
 exportClasses("Column")
 
@@ -129,6 +130,7 @@ exportMethods("%in%",
               "count",
               "countDistinct",
               "crc32",
+              "hash",
               "cume_dist",
               "date_add",
               "date_format",
@@ -274,6 +276,7 @@ export("as.DataFrame",
        "parquetFile",
        "read.df",
        "read.parquet",
+       "read.text",
        "sql",
        "table",
        "tableNames",

R/pkg/R/DataFrame.R

@@ -458,7 +458,10 @@ setMethod("registerTempTable",
 setMethod("insertInto",
           signature(x = "DataFrame", tableName = "character"),
           function(x, tableName, overwrite = FALSE) {
-            callJMethod(x@sdf, "insertInto", tableName, overwrite)
+            jmode <- convertToJSaveMode(ifelse(overwrite, "overwrite", "append"))
+            write <- callJMethod(x@sdf, "write")
+            write <- callJMethod(write, "mode", jmode)
+            callJMethod(write, "insertInto", tableName)
           })
 
 #' Cache
@@ -661,6 +664,34 @@ setMethod("saveAsParquetFile",
             write.parquet(x, path)
           })
 
+#' write.text
+#'
+#' Saves the content of the DataFrame in a text file at the specified path.
+#' The DataFrame must have only one column of string type with the name "value".
+#' Each row becomes a new line in the output file.
+#'
+#' @param x A SparkSQL DataFrame
+#' @param path The directory where the file is saved
+#'
+#' @family DataFrame functions
+#' @rdname write.text
+#' @name write.text
+#' @export
+#' @examples
+#'\dontrun{
+#' sc <- sparkR.init()
+#' sqlContext <- sparkRSQL.init(sc)
+#' path <- "path/to/file.txt"
+#' df <- read.text(sqlContext, path)
+#' write.text(df, "/tmp/sparkr-tmp/")
+#'}
+setMethod("write.text",
+          signature(x = "DataFrame", path = "character"),
+          function(x, path) {
+            write <- callJMethod(x@sdf, "write")
+            invisible(callJMethod(write, "text", path))
+          })
+
 #' Distinct
 #'
 #' Return a new DataFrame containing the distinct rows in this DataFrame.
@@ -1948,18 +1979,15 @@ setMethod("write.df",
               source <- callJMethod(sqlContext, "getConf", "spark.sql.sources.default",
                                     "org.apache.spark.sql.parquet")
             }
-            allModes <- c("append", "overwrite", "error", "ignore")
-            # nolint start
-            if (!(mode %in% allModes)) {
-              stop('mode should be one of "append", "overwrite", "error", "ignore"')
-            }
-            # nolint end
-            jmode <- callJStatic("org.apache.spark.sql.api.r.SQLUtils", "saveMode", mode)
+            jmode <- convertToJSaveMode(mode)
             options <- varargsToEnv(...)
             if (!is.null(path)) {
                 options[["path"]] <- path
             }
-            callJMethod(df@sdf, "save", source, jmode, options)
+            write <- callJMethod(df@sdf, "write")
+            write <- callJMethod(write, "format", source)
+            write <- callJMethod(write, "mode", jmode)
+            write <- callJMethod(write, "save", path)
           })
 
 #' @rdname write.df
@@ -2013,15 +2041,14 @@ setMethod("saveAsTable",
               source <- callJMethod(sqlContext, "getConf", "spark.sql.sources.default",
                                     "org.apache.spark.sql.parquet")
             }
-            allModes <- c("append", "overwrite", "error", "ignore")
-            # nolint start
-            if (!(mode %in% allModes)) {
-              stop('mode should be one of "append", "overwrite", "error", "ignore"')
-            }
-            # nolint end
-            jmode <- callJStatic("org.apache.spark.sql.api.r.SQLUtils", "saveMode", mode)
+            jmode <- convertToJSaveMode(mode)
             options <- varargsToEnv(...)
-            callJMethod(df@sdf, "saveAsTable", tableName, source, jmode, options)
+
+            write <- callJMethod(df@sdf, "write")
+            write <- callJMethod(write, "format", source)
+            write <- callJMethod(write, "mode", jmode)
+            write <- callJMethod(write, "options", options)
+            callJMethod(write, "saveAsTable", tableName)
           })
 
 #' summary