Merge remote-tracking branch 'upstream/master' into SPARK-3278

Author: zapletal-martin

.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

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

@@ -146,4 +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-submit

@@ -38,11 +38,19 @@ 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="$SPARK_HOME/conf/spark-defaults.conf"
+if [ -z "$SPARK_CONF_DIR" ]; then
+  export SPARK_CONF_DIR="$SPARK_HOME/conf"
+fi
+DEFAULT_PROPERTIES_FILE="$SPARK_CONF_DIR/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

@@ -24,7 +24,11 @@ set ORIG_ARGS=%*
 
 rem Reset the values of all variables used
 set SPARK_SUBMIT_DEPLOY_MODE=client
-set SPARK_SUBMIT_PROPERTIES_FILE=%SPARK_HOME%\conf\spark-defaults.conf
+
+if not defined %SPARK_CONF_DIR% (
+  set SPARK_CONF_DIR=%SPARK_HOME%\conf
+)
+set SPARK_SUBMIT_PROPERTIES_FILE=%SPARK_CONF_DIR%\spark-defaults.conf
 set SPARK_SUBMIT_DRIVER_MEMORY=
 set SPARK_SUBMIT_LIBRARY_PATH=
 set SPARK_SUBMIT_CLASSPATH=
@@ -45,11 +49,17 @@ 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

build/mvn

@@ -0,0 +1,149 @@
+#!/usr/bin/env bash
+
+#
+# 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.
+#
+
+# Determine the current working directory
+_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"
+# Preserve the calling directory
+_CALLING_DIR="$(pwd)"
+
+# Installs any application tarball given a URL, the expected tarball name,
+# and, optionally, a checkable binary path to determine if the binary has
+# already been installed
+## Arg1 - URL
+## Arg2 - Tarball Name
+## Arg3 - Checkable Binary
+install_app() {
+  local remote_tarball="$1/$2"
+  local local_tarball="${_DIR}/$2"
+  local binary="${_DIR}/$3"
+
+  # setup `curl` and `wget` silent options if we're running on Jenkins
+  local curl_opts=""
+  local wget_opts=""
+  if [ -n "$AMPLAB_JENKINS" ]; then
+    curl_opts="-s"
+    wget_opts="--quiet"
+  else
+    curl_opts="--progress-bar"
+    wget_opts="--progress=bar:force"
+  fi
+
+  if [ -z "$3" -o ! -f "$binary" ]; then
+    # check if we already have the tarball
+    # check if we have curl installed
+    # download application
+    [ ! -f "${local_tarball}" ] && [ -n "`which curl 2>/dev/null`" ] && \
+      echo "exec: curl ${curl_opts} ${remote_tarball}" && \
+      curl ${curl_opts} "${remote_tarball}" > "${local_tarball}"
+    # if the file still doesn't exist, lets try `wget` and cross our fingers
+    [ ! -f "${local_tarball}" ] && [ -n "`which wget 2>/dev/null`" ] && \
+      echo "exec: wget ${wget_opts} ${remote_tarball}" && \
+      wget ${wget_opts} -O "${local_tarball}" "${remote_tarball}"
+    # if both were unsuccessful, exit
+    [ ! -f "${local_tarball}" ] && \
+      echo -n "ERROR: Cannot download $2 with cURL or wget; " && \
+      echo "please install manually and try again." && \
+      exit 2
+    cd "${_DIR}" && tar -xzf "$2"
+    rm -rf "$local_tarball"
+  fi
+}
+
+# Install maven under the build/ folder
+install_mvn() {
+  install_app \
+    "http://apache.claz.org/maven/maven-3/3.2.3/binaries" \
+    "apache-maven-3.2.3-bin.tar.gz" \
+    "apache-maven-3.2.3/bin/mvn"
+  MVN_BIN="${_DIR}/apache-maven-3.2.3/bin/mvn"
+}
+
+# Install zinc under the build/ folder
+install_zinc() {
+  local zinc_path="zinc-0.3.5.3/bin/zinc"
+  [ ! -f "${zinc_path}" ] && ZINC_INSTALL_FLAG=1
+  install_app \
+    "http://downloads.typesafe.com/zinc/0.3.5.3" \
+    "zinc-0.3.5.3.tgz" \
+    "${zinc_path}"
+  ZINC_BIN="${_DIR}/${zinc_path}"
+}
+
+# Determine the Scala version from the root pom.xml file, set the Scala URL,
+# and, with that, download the specific version of Scala necessary under
+# the build/ folder
+install_scala() {
+  # determine the Scala version used in Spark
+  local scala_version=`grep "scala.version" "${_DIR}/../pom.xml" | \
+                       head -1 | cut -f2 -d'>' | cut -f1 -d'<'`
+  local scala_bin="${_DIR}/scala-${scala_version}/bin/scala"
+
+  install_app \
+    "http://downloads.typesafe.com/scala/${scala_version}" \
+    "scala-${scala_version}.tgz" \
+    "scala-${scala_version}/bin/scala"
+
+  SCALA_COMPILER="$(cd "$(dirname ${scala_bin})/../lib" && pwd)/scala-compiler.jar"
+  SCALA_LIBRARY="$(cd "$(dirname ${scala_bin})/../lib" && pwd)/scala-library.jar"
+}
+
+# Determines if a given application is already installed. If not, will attempt
+# to install
+## Arg1 - application name
+## Arg2 - Alternate path to local install under build/ dir
+check_and_install_app() {
+  # create the local environment variable in uppercase
+  local app_bin="`echo $1 | awk '{print toupper(\$0)}'`_BIN"
+  # some black magic to set the generated app variable (i.e. MVN_BIN) into the
+  # environment
+  eval "${app_bin}=`which $1 2>/dev/null`"
+
+  if [ -z "`which $1 2>/dev/null`" ]; then
+    install_$1
+  fi
+}
+
+# Setup healthy defaults for the Zinc port if none were provided from
+# the environment
+ZINC_PORT=${ZINC_PORT:-"3030"}
+
+# Check and install all applications necessary to build Spark
+check_and_install_app "mvn"
+
+# Install the proper version of Scala and Zinc for the build
+install_zinc
+install_scala
+
+# Reset the current working directory
+cd "${_CALLING_DIR}"
+
+# Now that zinc is ensured to be installed, check its status and, if its
+# not running or just installed, start it
+if [ -n "${ZINC_INSTALL_FLAG}" -o -z "`${ZINC_BIN} -status`" ]; then
+  ${ZINC_BIN} -shutdown
+  ${ZINC_BIN} -start -port ${ZINC_PORT} \
+    -scala-compiler "${SCALA_COMPILER}" \
+    -scala-library "${SCALA_LIBRARY}" &>/dev/null
+fi
+
+# Set any `mvn` options if not already present
+export MAVEN_OPTS=${MAVEN_OPTS:-"-Xmx2g -XX:MaxPermSize=512M -XX:ReservedCodeCacheSize=512m"}
+
+# Last, call the `mvn` command as usual
+${MVN_BIN} "$@"