Merge remote-tracking branch 'upstream/master' into SPARK-733

.gitignore

@@ -8,16 +8,19 @@
 *.pyc
 .idea/
 .idea_modules/
-sbt/*.jar
+build/*.jar
 .settings
 .cache
+cache
 .generated-mima*
-/build/
 work/
 out/
 .DS_Store
 third_party/libmesos.so
 third_party/libmesos.dylib
+build/apache-maven*
+build/zinc*
+build/scala*
 conf/java-opts
 conf/*.sh
 conf/*.cmd
@@ -51,10 +54,11 @@ checkpoint
 derby.log
 dist/
 dev/create-release/*txt
-dev/create-release/*new
+dev/create-release/*final
 spark-*-bin-*.tgz
 unit-tests.log
 /lib/
+ec2/lib/
 rat-results.txt
 scalastyle.txt
 scalastyle-output.xml

.rat-excludes

@@ -64,3 +64,4 @@ dist/*
 logs
 .*scalastyle-output.xml
 .*dependency-reduced-pom.xml
+known_translations

README.md

@@ -26,7 +26,7 @@ To build Spark and its example programs, run:
 
 (You do not need to do this if you downloaded a pre-built package.)
 More detailed documentation is available from the project site, at
-["Building Spark with Maven"](http://spark.apache.org/docs/latest/building-with-maven.html).
+["Building Spark with Maven"](http://spark.apache.org/docs/latest/building-spark.html).
 
 ## Interactive Scala Shell
 

assembly/pom.xml

@@ -354,5 +354,25 @@
         </dependency>
       </dependencies>
     </profile>
+
+    <!-- Profiles that disable inclusion of certain dependencies. -->
+    <profile>
+      <id>hadoop-provided</id>
+      <properties>
+        <hadoop.deps.scope>provided</hadoop.deps.scope>
+      </properties>
+    </profile>
+    <profile>
+      <id>hive-provided</id>
+      <properties>
+        <hive.deps.scope>provided</hive.deps.scope>
+      </properties>
+    </profile>
+    <profile>
+      <id>parquet-provided</id>
+      <properties>
+        <parquet.deps.scope>provided</parquet.deps.scope>
+      </properties>
+    </profile>
   </profiles>
 </project>

bagel/pom.xml

@@ -40,15 +40,6 @@
       <artifactId>spark-core_${scala.binary.version}</artifactId>
       <version>${project.version}</version>
     </dependency>
-    <dependency>
-      <groupId>org.eclipse.jetty</groupId>
-      <artifactId>jetty-server</artifactId>
-    </dependency>
-    <dependency>
-      <groupId>org.scalatest</groupId>
-      <artifactId>scalatest_${scala.binary.version}</artifactId>
-      <scope>test</scope>
-    </dependency>
     <dependency>
       <groupId>org.scalacheck</groupId>
       <artifactId>scalacheck_${scala.binary.version}</artifactId>
@@ -58,11 +49,5 @@
   <build>
     <outputDirectory>target/scala-${scala.binary.version}/classes</outputDirectory>
     <testOutputDirectory>target/scala-${scala.binary.version}/test-classes</testOutputDirectory>
-    <plugins>
-      <plugin>
-        <groupId>org.scalatest</groupId>
-        <artifactId>scalatest-maven-plugin</artifactId>
-      </plugin>
-    </plugins>
   </build>
 </project>

bagel/src/test/resources/log4j.properties

@@ -15,10 +15,10 @@
 # limitations under the License.
 #
 
-# Set everything to be logged to the file bagel/target/unit-tests.log
+# Set everything to be logged to the file target/unit-tests.log
 log4j.rootCategory=INFO, file
 log4j.appender.file=org.apache.log4j.FileAppender
-log4j.appender.file.append=false
+log4j.appender.file.append=true
 log4j.appender.file.file=target/unit-tests.log
 log4j.appender.file.layout=org.apache.log4j.PatternLayout
 log4j.appender.file.layout.ConversionPattern=%d{yy/MM/dd HH:mm:ss.SSS} %t %p %c{1}: %m%n

bin/compute-classpath.cmd

@@ -109,6 +109,13 @@ if "x%YARN_CONF_DIR%"=="x" goto no_yarn_conf_dir
   set CLASSPATH=%CLASSPATH%;%YARN_CONF_DIR%
 :no_yarn_conf_dir
 
+rem To allow for distributions to append needed libraries to the classpath (e.g. when
+rem using the "hadoop-provided" profile to build Spark), check SPARK_DIST_CLASSPATH and
+rem append it to tbe final classpath.
+if not "x%$SPARK_DIST_CLASSPATH%"=="x" (
+  set CLASSPATH=%CLASSPATH%;%SPARK_DIST_CLASSPATH%
+)
+
 rem A bit of a hack to allow calling this script within run2.cmd without seeing output
 if "%DONT_PRINT_CLASSPATH%"=="1" goto exit
 

bin/compute-classpath.sh

@@ -25,7 +25,11 @@ FWDIR="$(cd "`dirname "$0"`"/..; pwd)"
 
 . "$FWDIR"/bin/load-spark-env.sh
 
-CLASSPATH="$SPARK_CLASSPATH:$SPARK_SUBMIT_CLASSPATH"
+if [ -n "$SPARK_CLASSPATH" ]; then
+  CLASSPATH="$SPARK_CLASSPATH:$SPARK_SUBMIT_CLASSPATH"
+else
+  CLASSPATH="$SPARK_SUBMIT_CLASSPATH"
+fi
 
 # Build up classpath
 if [ -n "$SPARK_CONF_DIR" ]; then
@@ -142,4 +146,11 @@ if [ -n "$YARN_CONF_DIR" ]; then
   CLASSPATH="$CLASSPATH:$YARN_CONF_DIR"
 fi
 
+# To allow for distributions to append needed libraries to the classpath (e.g. when
+# using the "hadoop-provided" profile to build Spark), check SPARK_DIST_CLASSPATH and
+# append it to tbe final classpath.
+if [ -n "$SPARK_DIST_CLASSPATH" ]; then
+  CLASSPATH="$CLASSPATH:$SPARK_DIST_CLASSPATH"
+fi
+
 echo "$CLASSPATH"

bin/spark-shell2.cmd

@@ -19,4 +19,23 @@ rem
 
 set SPARK_HOME=%~dp0..
 
-cmd /V /E /C %SPARK_HOME%\bin\spark-submit.cmd --class org.apache.spark.repl.Main %* spark-shell
+echo "%*" | findstr " --help -h" >nul
+if %ERRORLEVEL% equ 0 (
+  call :usage
+  exit /b 0
+)
+
+call %SPARK_HOME%\bin\windows-utils.cmd %*
+if %ERRORLEVEL% equ 1 (
+  call :usage
+  exit /b 1
+)
+
+cmd /V /E /C %SPARK_HOME%\bin\spark-submit.cmd --class org.apache.spark.repl.Main %SUBMISSION_OPTS% spark-shell %APPLICATION_OPTS%
+
+exit /b 0
+
+:usage
+echo "Usage: .\bin\spark-shell.cmd [options]" >&2
+%SPARK_HOME%\bin\spark-submit --help 2>&1 | findstr /V "Usage" 1>&2
+exit /b 0

bin/spark-submit

@@ -38,11 +38,16 @@ while (($#)); do
     export SPARK_SUBMIT_CLASSPATH=$2
   elif [ "$1" = "--driver-java-options" ]; then
     export SPARK_SUBMIT_OPTS=$2
+  elif [ "$1" = "--master" ]; then
+    export MASTER=$2
   fi
   shift
 done
 
 DEFAULT_PROPERTIES_FILE="$SPARK_HOME/conf/spark-defaults.conf"
+if [ "$MASTER" == "yarn-cluster" ]; then
+  SPARK_SUBMIT_DEPLOY_MODE=cluster
+fi
 export SPARK_SUBMIT_DEPLOY_MODE=${SPARK_SUBMIT_DEPLOY_MODE:-"client"}
 export SPARK_SUBMIT_PROPERTIES_FILE=${SPARK_SUBMIT_PROPERTIES_FILE:-"$DEFAULT_PROPERTIES_FILE"}
 

bin/spark-submit2.cmd

@@ -45,11 +45,17 @@ if [%1] == [] goto continue
     set SPARK_SUBMIT_CLASSPATH=%2
   ) else if [%1] == [--driver-java-options] (
     set SPARK_SUBMIT_OPTS=%2
+  ) else if [%1] == [--master] (
+    set MASTER=%2
   )
   shift
 goto loop
 :continue
 
+if [%MASTER%] == [yarn-cluster] (
+  set SPARK_SUBMIT_DEPLOY_MODE=cluster
+)
+
 rem For client mode, the driver will be launched in the same JVM that launches
 rem SparkSubmit, so we may need to read the properties file for any extra class
 rem paths, library paths, java options and memory early on. Otherwise, it will

bin/windows-utils.cmd

@@ -0,0 +1,59 @@
+rem
+rem Licensed to the Apache Software Foundation (ASF) under one or more
+rem contributor license agreements.  See the NOTICE file distributed with
+rem this work for additional information regarding copyright ownership.
+rem The ASF licenses this file to You under the Apache License, Version 2.0
+rem (the "License"); you may not use this file except in compliance with
+rem the License.  You may obtain a copy of the License at
+rem
+rem    http://www.apache.org/licenses/LICENSE-2.0
+rem
+rem Unless required by applicable law or agreed to in writing, software
+rem distributed under the License is distributed on an "AS IS" BASIS,
+rem WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+rem See the License for the specific language governing permissions and
+rem limitations under the License.
+rem
+
+rem Gather all spark-submit options into SUBMISSION_OPTS
+
+set SUBMISSION_OPTS=
+set APPLICATION_OPTS=
+
+rem NOTE: If you add or remove spark-sumbmit options,
+rem modify NOT ONLY this script but also SparkSubmitArgument.scala
+
+:OptsLoop
+if "x%1"=="x" (
+  goto :OptsLoopEnd
+)
+
+SET opts="\<--master\> \<--deploy-mode\> \<--class\> \<--name\> \<--jars\> \<--py-files\> \<--files\>"
+SET opts="%opts:~1,-1% \<--conf\> \<--properties-file\> \<--driver-memory\> \<--driver-java-options\>"
+SET opts="%opts:~1,-1% \<--driver-library-path\> \<--driver-class-path\> \<--executor-memory\>"
+SET opts="%opts:~1,-1% \<--driver-cores\> \<--total-executor-cores\> \<--executor-cores\> \<--queue\>"
+SET opts="%opts:~1,-1% \<--num-executors\> \<--archives\>"
+
+echo %1 | findstr %opts% >nul
+if %ERRORLEVEL% equ 0 (
+  if "x%2"=="x" (
+    echo "%1" requires an argument. >&2
+    exit /b 1
+  )
+  set SUBMISSION_OPTS=%SUBMISSION_OPTS% %1 %2
+  shift
+  shift
+  goto :OptsLoop
+)
+echo %1 | findstr "\<--verbose\> \<-v\> \<--supervise\>" >nul
+if %ERRORLEVEL% equ 0 (
+  set SUBMISSION_OPTS=%SUBMISSION_OPTS% %1
+  shift
+  goto :OptsLoop
+)
+set APPLICATION_OPTS=%APPLICATION_OPTS% %1
+shift
+goto :OptsLoop
+
+:OptsLoopEnd
+exit /b 0