Merge branch 'master' of https://github.com/apache/spark into SPARK-1…

…8408-yunn-api-improvements

.github/PULL_REQUEST_TEMPLATE

@@ -7,4 +7,4 @@
 (Please explain how this patch was tested. E.g. unit tests, integration tests, manual tests)
 (If this patch involves UI changes, please attach a screenshot; otherwise, remove this)
 
-Please review https://cwiki.apache.org/confluence/display/SPARK/Contributing+to+Spark before opening a pull request.
+Please review http://spark.apache.org/contributing.html before opening a pull request.

CONTRIBUTING.md

@@ -1,12 +1,12 @@
 ## Contributing to Spark
 
 *Before opening a pull request*, review the 
-[Contributing to Spark wiki](https://cwiki.apache.org/confluence/display/SPARK/Contributing+to+Spark). 
+[Contributing to Spark guide](http://spark.apache.org/contributing.html). 
 It lists steps that are required before creating a PR. In particular, consider:
 
 - Is the change important and ready enough to ask the community to spend time reviewing?
 - Have you searched for existing, related JIRAs and pull requests?
-- Is this a new feature that can stand alone as a [third party project](https://cwiki.apache.org/confluence/display/SPARK/Third+Party+Projects) ?
+- Is this a new feature that can stand alone as a [third party project](http://spark.apache.org/third-party-projects.html) ?
 - Is the change being proposed clearly explained and motivated?
 
 When you contribute code, you affirm that the contribution is your original work and that you 

R/README.md

@@ -51,7 +51,7 @@ sparkR.session()
 
 #### Making changes to SparkR
 
-The [instructions](https://cwiki.apache.org/confluence/display/SPARK/Contributing+to+Spark) for making contributions to Spark also apply to SparkR.
+The [instructions](http://spark.apache.org/contributing.html) for making contributions to Spark also apply to SparkR.
 If you only make R file changes (i.e. no Scala changes) then you can just re-install the R package using `R/install-dev.sh` and test your changes.
 Once you have made your changes, please include unit tests for them and run existing unit tests using the `R/run-tests.sh` script as described below.
 

R/pkg/DESCRIPTION

@@ -11,7 +11,7 @@ Authors@R: c(person("Shivaram", "Venkataraman", role = c("aut", "cre"),
                     email = "felixcheung@apache.org"),
              person(family = "The Apache Software Foundation", role = c("aut", "cph")))
 URL: http://www.apache.org/ http://spark.apache.org/
-BugReports: https://cwiki.apache.org/confluence/display/SPARK/Contributing+to+Spark#ContributingtoSpark-ContributingBugReports
+BugReports: http://spark.apache.org/contributing.html
 Depends:
     R (>= 3.0),
     methods

R/pkg/R/DataFrame.R

@@ -2541,7 +2541,8 @@ generateAliasesForIntersectedCols <- function (x, intersectedColNames, suffix) {
 #'
 #' Return a new SparkDataFrame containing the union of rows in this SparkDataFrame
 #' and another SparkDataFrame. This is equivalent to \code{UNION ALL} in SQL.
-#' Note that this does not remove duplicate rows across the two SparkDataFrames.
+#'
+#' Note: This does not remove duplicate rows across the two SparkDataFrames.
 #'
 #' @param x A SparkDataFrame
 #' @param y A SparkDataFrame
@@ -2584,7 +2585,8 @@ setMethod("unionAll",
 #' Union two or more SparkDataFrames
 #'
 #' Union two or more SparkDataFrames. This is equivalent to \code{UNION ALL} in SQL.
-#' Note that this does not remove duplicate rows across the two SparkDataFrames.
+#'
+#' Note: This does not remove duplicate rows across the two SparkDataFrames.
 #'
 #' @param x a SparkDataFrame.
 #' @param ... additional SparkDataFrame(s).

R/pkg/R/functions.R

@@ -2296,7 +2296,7 @@ setMethod("n", signature(x = "Column"),
 #' A pattern could be for instance \preformatted{dd.MM.yyyy} and could return a string like '18.03.1993'. All
 #' pattern letters of \code{java.text.SimpleDateFormat} can be used.
 #'
-#' NOTE: Use when ever possible specialized functions like \code{year}. These benefit from a
+#' Note: Use when ever possible specialized functions like \code{year}. These benefit from a
 #' specialized implementation.
 #'
 #' @param y Column to compute on.
@@ -2341,7 +2341,7 @@ setMethod("from_utc_timestamp", signature(y = "Column", x = "character"),
 #' Locate the position of the first occurrence of substr column in the given string.
 #' Returns null if either of the arguments are null.
 #'
-#' NOTE: The position is not zero based, but 1 based index. Returns 0 if substr
+#' Note: The position is not zero based, but 1 based index. Returns 0 if substr
 #' could not be found in str.
 #'
 #' @param y column to check
@@ -2779,7 +2779,8 @@ setMethod("window", signature(x = "Column"),
 #' locate
 #'
 #' Locate the position of the first occurrence of substr.
-#' NOTE: The position is not zero based, but 1 based index. Returns 0 if substr
+#'
+#' Note: The position is not zero based, but 1 based index. Returns 0 if substr
 #' could not be found in str.
 #'
 #' @param substr a character string to be matched.

R/pkg/R/mllib.R

@@ -278,8 +278,10 @@ setMethod("glm", signature(formula = "formula", family = "ANY", data = "SparkDat
 
 #' @param object a fitted generalized linear model.
 #' @return \code{summary} returns a summary object of the fitted model, a list of components
-#'         including at least the coefficients, null/residual deviance, null/residual degrees
-#'         of freedom, AIC and number of iterations IRLS takes.
+#'         including at least the coefficients matrix (which includes coefficients, standard error
+#'         of coefficients, t value and p value), null/residual deviance, null/residual degrees of
+#'         freedom, AIC and number of iterations IRLS takes. If there are collinear columns
+#'         in you data, the coefficients matrix only provides coefficients.
 #'
 #' @rdname spark.glm
 #' @export
@@ -303,9 +305,18 @@ setMethod("summary", signature(object = "GeneralizedLinearRegressionModel"),
             } else {
               dataFrame(callJMethod(jobj, "rDevianceResiduals"))
             }
-            coefficients <- matrix(coefficients, ncol = 4)
-            colnames(coefficients) <- c("Estimate", "Std. Error", "t value", "Pr(>|t|)")
-            rownames(coefficients) <- unlist(features)
+            # If the underlying WeightedLeastSquares using "normal" solver, we can provide
+            # coefficients, standard error of coefficients, t value and p value. Otherwise,
+            # it will be fitted by local "l-bfgs", we can only provide coefficients.
+            if (length(features) == length(coefficients)) {
+              coefficients <- matrix(coefficients, ncol = 1)
+              colnames(coefficients) <- c("Estimate")
+              rownames(coefficients) <- unlist(features)
+            } else {
+              coefficients <- matrix(coefficients, ncol = 4)
+              colnames(coefficients) <- c("Estimate", "Std. Error", "t value", "Pr(>|t|)")
+              rownames(coefficients) <- unlist(features)
+            }
             ans <- list(deviance.resid = deviance.resid, coefficients = coefficients,
                         dispersion = dispersion, null.deviance = null.deviance,
                         deviance = deviance, df.null = df.null, df.residual = df.residual,

R/pkg/R/sparkR.R

@@ -373,8 +373,13 @@ sparkR.session <- function(
     overrideEnvs(sparkConfigMap, paramMap)
   }
 
+  deployMode <- ""
+  if (exists("spark.submit.deployMode", envir = sparkConfigMap)) {
+    deployMode <- sparkConfigMap[["spark.submit.deployMode"]]
+  }
+
   if (!exists(".sparkRjsc", envir = .sparkREnv)) {
-    retHome <- sparkCheckInstall(sparkHome, master)
+    retHome <- sparkCheckInstall(sparkHome, master, deployMode)
     if (!is.null(retHome)) sparkHome <- retHome
     sparkExecutorEnvMap <- new.env()
     sparkR.sparkContext(master, appName, sparkHome, sparkConfigMap, sparkExecutorEnvMap,
@@ -550,24 +555,27 @@ processSparkPackages <- function(packages) {
 #
 # @param sparkHome directory to find Spark package.
 # @param master the Spark master URL, used to check local or remote mode.
+# @param deployMode whether to deploy your driver on the worker nodes (cluster)
+#        or locally as an external client (client).
 # @return NULL if no need to update sparkHome, and new sparkHome otherwise.
-sparkCheckInstall <- function(sparkHome, master) {
+sparkCheckInstall <- function(sparkHome, master, deployMode) {
   if (!isSparkRShell()) {
     if (!is.na(file.info(sparkHome)$isdir)) {
       msg <- paste0("Spark package found in SPARK_HOME: ", sparkHome)
       message(msg)
       NULL
     } else {
-      if (!nzchar(master) || isMasterLocal(master)) {
-        msg <- paste0("Spark not found in SPARK_HOME: ",
-                      sparkHome)
+      if (isMasterLocal(master)) {
+        msg <- paste0("Spark not found in SPARK_HOME: ", sparkHome)
         message(msg)
         packageLocalDir <- install.spark()
         packageLocalDir
-      } else {
+      } else if (isClientMode(master) || deployMode == "client") {
         msg <- paste0("Spark not found in SPARK_HOME: ",
                       sparkHome, "\n", installInstruction("remote"))
         stop(msg)
+      } else {
+        NULL
       }
     }
   } else {

R/pkg/R/utils.R

@@ -777,6 +777,10 @@ isMasterLocal <- function(master) {
   grepl("^local(\\[([0-9]+|\\*)\\])?$", master, perl = TRUE)
 }
 
+isClientMode <- function(master) {
+  grepl("([a-z]+)-client$", master, perl = TRUE)
+}
+
 isSparkRShell <- function() {
   grepl(".*shell\\.R$", Sys.getenv("R_PROFILE_USER"), perl = TRUE)
 }

R/pkg/inst/tests/testthat/test_mllib.R

@@ -169,6 +169,15 @@ test_that("spark.glm summary", {
   df <- suppressWarnings(createDataFrame(data))
   regStats <- summary(spark.glm(df, b ~ a1 + a2, regParam = 1.0))
   expect_equal(regStats$aic, 14.00976, tolerance = 1e-4) # 14.00976 is from summary() result
+
+  # Test spark.glm works on collinear data
+  A <- matrix(c(1, 2, 3, 4, 2, 4, 6, 8), 4, 2)
+  b <- c(1, 2, 3, 4)
+  data <- as.data.frame(cbind(A, b))
+  df <- createDataFrame(data)
+  stats <- summary(spark.glm(df, b ~ . - 1))
+  coefs <- unlist(stats$coefficients)
+  expect_true(all(abs(c(0.5, 0.25) - coefs) < 1e-4))
 })
 
 test_that("spark.glm save/load", {

R/pkg/inst/tests/testthat/test_sparkR.R

@@ -0,0 +1,46 @@
+#
+# Licensed to the Apache Software Foundation (ASF) under one or more
+# contributor license agreements.  See the NOTICE file distributed with
+# this work for additional information regarding copyright ownership.
+# The ASF licenses this file to You under the Apache License, Version 2.0
+# (the "License"); you may not use this file except in compliance with
+# the License.  You may obtain a copy of the License at
+#
+#    http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+
+context("functions in sparkR.R")
+
+test_that("sparkCheckInstall", {
+  # "local, yarn-client, mesos-client" mode, SPARK_HOME was set correctly,
+  # and the SparkR job was submitted by "spark-submit"
+  sparkHome <- paste0(tempdir(), "/", "sparkHome")
+  dir.create(sparkHome)
+  master <- ""
+  deployMode <- ""
+  expect_true(is.null(sparkCheckInstall(sparkHome, master, deployMode)))
+  unlink(sparkHome, recursive = TRUE)
+
+  # "yarn-cluster, mesos-cluster" mode, SPARK_HOME was not set,
+  # and the SparkR job was submitted by "spark-submit"
+  sparkHome <- ""
+  master <- ""
+  deployMode <- ""
+  expect_true(is.null(sparkCheckInstall(sparkHome, master, deployMode)))
+
+  # "yarn-client, mesos-client" mode, SPARK_HOME was not set
+  sparkHome <- ""
+  master <- "yarn-client"
+  deployMode <- ""
+  expect_error(sparkCheckInstall(sparkHome, master, deployMode))
+  sparkHome <- ""
+  master <- ""
+  deployMode <- "client"
+  expect_error(sparkCheckInstall(sparkHome, master, deployMode))
+})

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

@@ -2684,7 +2684,7 @@ test_that("Call DataFrameWriter.load() API in Java without path and check argume
   # It makes sure that we can omit path argument in read.df API and then it calls
   # DataFrameWriter.load() without path.
   expect_error(read.df(source = "json"),
-               paste("Error in loadDF : analysis error - Unable to infer schema for JSON at .",
+               paste("Error in loadDF : analysis error - Unable to infer schema for JSON.",
                      "It must be specified manually"))
   expect_error(read.df("arbitrary_path"), "Error in loadDF : analysis error - Path does not exist")
   expect_error(read.json("arbitrary_path"), "Error in json : analysis error - Path does not exist")

README.md

@@ -29,8 +29,9 @@ To build Spark and its example programs, run:
 You can build Spark using more than one thread by using the -T option with Maven, see ["Parallel builds in Maven 3"](https://cwiki.apache.org/confluence/display/MAVEN/Parallel+builds+in+Maven+3).
 More detailed documentation is available from the project site, at
 ["Building Spark"](http://spark.apache.org/docs/latest/building-spark.html).
-For developing Spark using an IDE, see [Eclipse](https://cwiki.apache.org/confluence/display/SPARK/Useful+Developer+Tools#UsefulDeveloperTools-Eclipse)
-and [IntelliJ](https://cwiki.apache.org/confluence/display/SPARK/Useful+Developer+Tools#UsefulDeveloperTools-IntelliJ).
+
+For general development tips, including info on developing Spark using an IDE, see 
+[http://spark.apache.org/developer-tools.html](the Useful Developer Tools page).
 
 ## Interactive Scala Shell
 
@@ -80,7 +81,7 @@ can be run using:
     ./dev/run-tests
 
 Please see the guidance on how to
-[run tests for a module, or individual tests](https://cwiki.apache.org/confluence/display/SPARK/Useful+Developer+Tools).
+[run tests for a module, or individual tests](http://spark.apache.org/developer-tools.html#individual-tests).
 
 ## A Note About Hadoop Versions
 
@@ -100,5 +101,5 @@ in the online documentation for an overview on how to configure Spark.
 
 ## Contributing
 
-Please review the [Contribution to Spark](https://cwiki.apache.org/confluence/display/SPARK/Contributing+to+Spark)
-wiki for information on how to get started contributing to the project.
+Please review the [Contribution to Spark guide](http://spark.apache.org/contributing.html)
+for information on how to get started contributing to the project.

core/src/main/java/org/apache/spark/memory/TaskMemoryManager.java

@@ -378,14 +378,14 @@ public long cleanUpAllAllocatedMemory() {
       for (MemoryConsumer c: consumers) {
         if (c != null && c.getUsed() > 0) {
           // In case of failed task, it's normal to see leaked memory
-          logger.warn("leak " + Utils.bytesToString(c.getUsed()) + " memory from " + c);
+          logger.debug("unreleased " + Utils.bytesToString(c.getUsed()) + " memory from " + c);
         }
       }
       consumers.clear();
 
       for (MemoryBlock page : pageTable) {
         if (page != null) {
-          logger.warn("leak a page: " + page + " in task " + taskAttemptId);
+          logger.debug("unreleased page: " + page + " in task " + taskAttemptId);
           memoryManager.tungstenMemoryAllocator().free(page);
         }
       }