diff --git a/.gitignore b/.gitignore index 20095dd973..9757054a50 100644 --- a/.gitignore +++ b/.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 diff --git a/build/mvn b/build/mvn new file mode 100755 index 0000000000..dde486a8ac --- /dev/null +++ b/build/mvn @@ -0,0 +1,132 @@ +#!/usr/bin/env bash + +# 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 +${MVN_BIN} "$@" diff --git a/build/sbt b/build/sbt new file mode 100755 index 0000000000..0a251d97db --- /dev/null +++ b/build/sbt @@ -0,0 +1,111 @@ +#!/usr/bin/env bash + +# When creating new tests for Spark SQL Hive, the HADOOP_CLASSPATH must contain the hive jars so +# that we can run Hive to generate the golden answer. This is not required for normal development +# or testing. +for i in "$HIVE_HOME"/lib/* +do HADOOP_CLASSPATH="$HADOOP_CLASSPATH:$i" +done +export HADOOP_CLASSPATH + +realpath () { +( + TARGET_FILE="$1" + + cd "$(dirname "$TARGET_FILE")" + TARGET_FILE="$(basename "$TARGET_FILE")" + + COUNT=0 + while [ -L "$TARGET_FILE" -a $COUNT -lt 100 ] + do + TARGET_FILE="$(readlink "$TARGET_FILE")" + cd $(dirname "$TARGET_FILE") + TARGET_FILE="$(basename $TARGET_FILE)" + COUNT=$(($COUNT + 1)) + done + + echo "$(pwd -P)/"$TARGET_FILE"" +) +} + +. "$(dirname "$(realpath "$0")")"/sbt-launch-lib.bash + + +declare -r noshare_opts="-Dsbt.global.base=project/.sbtboot -Dsbt.boot.directory=project/.boot -Dsbt.ivy.home=project/.ivy" +declare -r sbt_opts_file=".sbtopts" +declare -r etc_sbt_opts_file="/etc/sbt/sbtopts" + +usage() { + cat < path to global settings/plugins directory (default: ~/.sbt) + -sbt-boot path to shared boot directory (default: ~/.sbt/boot in 0.11 series) + -ivy path to local Ivy repository (default: ~/.ivy2) + -mem set memory options (default: $sbt_mem, which is $(get_mem_opts $sbt_mem)) + -no-share use all local caches; no sharing + -no-global uses global caches, but does not use global ~/.sbt directory. + -jvm-debug Turn on JVM debugging, open at the given port. + -batch Disable interactive mode + + # sbt version (default: from project/build.properties if present, else latest release) + -sbt-version use the specified version of sbt + -sbt-jar use the specified jar as the sbt launcher + -sbt-rc use an RC version of sbt + -sbt-snapshot use a snapshot version of sbt + + # java version (default: java from PATH, currently $(java -version 2>&1 | grep version)) + -java-home alternate JAVA_HOME + + # jvm options and output control + JAVA_OPTS environment variable, if unset uses "$java_opts" + SBT_OPTS environment variable, if unset uses "$default_sbt_opts" + .sbtopts if this file exists in the current directory, it is + prepended to the runner args + /etc/sbt/sbtopts if this file exists, it is prepended to the runner args + -Dkey=val pass -Dkey=val directly to the java runtime + -J-X pass option -X directly to the java runtime + (-J is stripped) + -S-X add -X to sbt's scalacOptions (-S is stripped) + -PmavenProfiles Enable a maven profile for the build. + +In the case of duplicated or conflicting options, the order above +shows precedence: JAVA_OPTS lowest, command line options highest. +EOM +} + +process_my_args () { + while [[ $# -gt 0 ]]; do + case "$1" in + -no-colors) addJava "-Dsbt.log.noformat=true" && shift ;; + -no-share) addJava "$noshare_opts" && shift ;; + -no-global) addJava "-Dsbt.global.base=$(pwd)/project/.sbtboot" && shift ;; + -sbt-boot) require_arg path "$1" "$2" && addJava "-Dsbt.boot.directory=$2" && shift 2 ;; + -sbt-dir) require_arg path "$1" "$2" && addJava "-Dsbt.global.base=$2" && shift 2 ;; + -debug-inc) addJava "-Dxsbt.inc.debug=true" && shift ;; + -batch) exec &1 \ | grep -e "^java version" --max-count=1 \ | sed "s/java version \"\(.*\)\.\(.*\)\.\(.*\)\"/\1\2/" ) - + if [ "$JAVA_VERSION" -lt 18 ]; then echo "[warn] Java 8 tests will not run because JDK version is < 1.8." fi @@ -79,7 +79,7 @@ export SBT_MAVEN_PROFILES_ARGS="$SBT_MAVEN_PROFILES_ARGS -Pkinesis-asl" # Partial solution for SPARK-1455. if [ -n "$AMPLAB_JENKINS" ]; then git fetch origin master:master - + sql_diffs=$( git diff --name-only master \ | grep -e "^sql/" -e "^bin/spark-sql" -e "^sbin/start-thriftserver.sh" @@ -93,7 +93,7 @@ if [ -n "$AMPLAB_JENKINS" ]; then if [ -n "$sql_diffs" ]; then echo "[info] Detected changes in SQL. Will run Hive test suite." _RUN_SQL_TESTS=true - + if [ -z "$non_sql_diffs" ]; then echo "[info] Detected no changes except in SQL. Will only run SQL tests." _SQL_TESTS_ONLY=true @@ -151,7 +151,7 @@ CURRENT_BLOCK=$BLOCK_BUILD HIVE_12_BUILD_ARGS="$SBT_MAVEN_PROFILES_ARGS -Phive -Phive-thriftserver -Phive-0.12.0" echo "[info] Compile with Hive 0.12.0" echo -e "q\n" \ - | sbt/sbt $HIVE_12_BUILD_ARGS clean hive/compile hive-thriftserver/compile \ + | build/sbt $HIVE_12_BUILD_ARGS clean hive/compile hive-thriftserver/compile \ | grep -v -e "info.*Resolving" -e "warn.*Merging" -e "info.*Including" # Then build with default Hive version (0.13.1) because tests are based on this version @@ -160,7 +160,7 @@ CURRENT_BLOCK=$BLOCK_BUILD echo "[info] Building Spark with these arguments: $SBT_MAVEN_PROFILES_ARGS"\ " -Phive -Phive-thriftserver" echo -e "q\n" \ - | sbt/sbt $SBT_MAVEN_PROFILES_ARGS -Phive -Phive-thriftserver package assembly/assembly \ + | build/sbt $SBT_MAVEN_PROFILES_ARGS -Phive -Phive-thriftserver package assembly/assembly \ | grep -v -e "info.*Resolving" -e "warn.*Merging" -e "info.*Including" } @@ -177,7 +177,7 @@ CURRENT_BLOCK=$BLOCK_SPARK_UNIT_TESTS if [ -n "$_RUN_SQL_TESTS" ]; then SBT_MAVEN_PROFILES_ARGS="$SBT_MAVEN_PROFILES_ARGS -Phive -Phive-thriftserver" fi - + if [ -n "$_SQL_TESTS_ONLY" ]; then # This must be an array of individual arguments. Otherwise, having one long string # will be interpreted as a single test, which doesn't work. @@ -185,19 +185,19 @@ CURRENT_BLOCK=$BLOCK_SPARK_UNIT_TESTS else SBT_MAVEN_TEST_ARGS=("test") fi - + echo "[info] Running Spark tests with these arguments: $SBT_MAVEN_PROFILES_ARGS ${SBT_MAVEN_TEST_ARGS[@]}" - + # NOTE: echo "q" is needed because sbt on encountering a build file with failure # (either resolution or compilation) prompts the user for input either q, r, etc # to quit or retry. This echo is there to make it not block. - # NOTE: Do not quote $SBT_MAVEN_PROFILES_ARGS or else it will be interpreted as a + # NOTE: Do not quote $SBT_MAVEN_PROFILES_ARGS or else it will be interpreted as a # single argument! # "${SBT_MAVEN_TEST_ARGS[@]}" is cool because it's an array. # QUESTION: Why doesn't 'yes "q"' work? # QUESTION: Why doesn't 'grep -v -e "^\[info\] Resolving"' work? echo -e "q\n" \ - | sbt/sbt $SBT_MAVEN_PROFILES_ARGS "${SBT_MAVEN_TEST_ARGS[@]}" \ + | build/sbt $SBT_MAVEN_PROFILES_ARGS "${SBT_MAVEN_TEST_ARGS[@]}" \ | grep -v -e "info.*Resolving" -e "warn.*Merging" -e "info.*Including" } diff --git a/dev/scalastyle b/dev/scalastyle index 3a4df6e4bf..86919227ed 100755 --- a/dev/scalastyle +++ b/dev/scalastyle @@ -17,9 +17,9 @@ # limitations under the License. # -echo -e "q\n" | sbt/sbt -Phive -Phive-thriftserver scalastyle > scalastyle.txt +echo -e "q\n" | build/sbt -Phive -Phive-thriftserver scalastyle > scalastyle.txt # Check style with YARN built too -echo -e "q\n" | sbt/sbt -Pyarn -Phadoop-2.2 -Dhadoop.version=2.2.0 scalastyle \ +echo -e "q\n" | build/sbt -Pyarn -Phadoop-2.2 -Dhadoop.version=2.2.0 scalastyle \ >> scalastyle.txt ERRORS=$(cat scalastyle.txt | awk '{if($1~/error/)print}') diff --git a/docs/README.md b/docs/README.md index 1194840380..8a54724c4b 100644 --- a/docs/README.md +++ b/docs/README.md @@ -21,7 +21,7 @@ read those text files directly if you want. Start with index.md. The markdown code can be compiled to HTML using the [Jekyll tool](http://jekyllrb.com). `Jekyll` and a few dependencies must be installed for this to work. We recommend -installing via the Ruby Gem dependency manager. Since the exact HTML output +installing via the Ruby Gem dependency manager. Since the exact HTML output varies between versions of Jekyll and its dependencies, we list specific versions here in some cases: @@ -60,7 +60,7 @@ We use Sphinx to generate Python API docs, so you will need to install it by run ## API Docs (Scaladoc and Sphinx) -You can build just the Spark scaladoc by running `sbt/sbt doc` from the SPARK_PROJECT_ROOT directory. +You can build just the Spark scaladoc by running `build/sbt doc` from the SPARK_PROJECT_ROOT directory. Similarly, you can build just the PySpark docs by running `make html` from the SPARK_PROJECT_ROOT/python/docs directory. Documentation is only generated for classes that are listed as @@ -68,7 +68,7 @@ public in `__init__.py`. When you run `jekyll` in the `docs` directory, it will also copy over the scaladoc for the various Spark subprojects into the `docs` directory (and then also into the `_site` directory). We use a -jekyll plugin to run `sbt/sbt doc` before building the site so if you haven't run it (recently) it +jekyll plugin to run `build/sbt doc` before building the site so if you haven't run it (recently) it may take some time as it generates all of the scaladoc. The jekyll plugin also generates the PySpark docs [Sphinx](http://sphinx-doc.org/). diff --git a/docs/_plugins/copy_api_dirs.rb b/docs/_plugins/copy_api_dirs.rb index 4566a2fff5..3c626a0b7f 100644 --- a/docs/_plugins/copy_api_dirs.rb +++ b/docs/_plugins/copy_api_dirs.rb @@ -25,8 +25,8 @@ if not (ENV['SKIP_API'] == '1' or ENV['SKIP_SCALADOC'] == '1') curr_dir = pwd cd("..") - puts "Running 'sbt/sbt -Pkinesis-asl compile unidoc' from " + pwd + "; this may take a few minutes..." - puts `sbt/sbt -Pkinesis-asl compile unidoc` + puts "Running 'build/sbt -Pkinesis-asl compile unidoc' from " + pwd + "; this may take a few minutes..." + puts `build/sbt -Pkinesis-asl compile unidoc` puts "Moving back into docs dir." cd("docs") diff --git a/docs/building-spark.md b/docs/building-spark.md index dab3d2aef4..c1bcd91b5b 100644 --- a/docs/building-spark.md +++ b/docs/building-spark.md @@ -9,6 +9,15 @@ redirect_from: "building-with-maven.html" Building Spark using Maven requires Maven 3.0.4 or newer and Java 6+. +# Building with `build/mvn` + +Spark now comes packaged with a self-contained Maven installation to ease building and deployment of Spark from source located under the `build/` directory. This script will automatically download and setup all necessary build requirements ([Maven](https://maven.apache.org/), [Scala](http://www.scala-lang.org/), and [Zinc](https://github.com/typesafehub/zinc)) locally within the `build/` directory itself. It honors any `mvn` binary if present already, however, will pull down its own copy of Scala and Zinc regardless to ensure proper version requirements are met. `build/mvn` execution acts as a pass through to the `mvn` call allowing easy transition from previous build methods. As an example, one can build a version of Spark as follows: + +{% highlight bash %} +build/mvn -Pyarn -Phadoop-2.4 -Dhadoop.version=2.4.0 -DskipTests clean package +{% endhighlight %} + +Other build examples can be found below. # Setting up Maven's Memory Usage @@ -28,7 +37,9 @@ If you don't run this, you may see errors like the following: You can fix this by setting the `MAVEN_OPTS` variable as discussed before. -**Note:** *For Java 8 and above this step is not required.* +**Note:** +* *For Java 8 and above this step is not required.* +* *If using `build/mvn` and `MAVEN_OPTS` were not already set, the script will automate this for you.* # Specifying the Hadoop Version @@ -84,7 +95,7 @@ mvn -Pyarn -Phadoop-2.3 -Dhadoop.version=2.3.0 -Dyarn.version=2.2.0 -DskipTests # Building With Hive and JDBC Support To enable Hive integration for Spark SQL along with its JDBC server and CLI, add the `-Phive` and `Phive-thriftserver` profiles to your existing build options. -By default Spark will build with Hive 0.13.1 bindings. You can also build for +By default Spark will build with Hive 0.13.1 bindings. You can also build for Hive 0.12.0 using the `-Phive-0.12.0` profile. {% highlight bash %} # Apache Hadoop 2.4.X with Hive 13 support @@ -106,7 +117,7 @@ supported in Scala 2.11 builds. # Spark Tests in Maven -Tests are run by default via the [ScalaTest Maven plugin](http://www.scalatest.org/user_guide/using_the_scalatest_maven_plugin). +Tests are run by default via the [ScalaTest Maven plugin](http://www.scalatest.org/user_guide/using_the_scalatest_maven_plugin). Some of the tests require Spark to be packaged first, so always run `mvn package` with `-DskipTests` the first time. The following is an example of a correct (build, test) sequence: @@ -124,7 +135,7 @@ We use the scala-maven-plugin which supports incremental and continuous compilat mvn scala:cc -should run continuous compilation (i.e. wait for changes). However, this has not been tested +should run continuous compilation (i.e. wait for changes). However, this has not been tested extensively. A couple of gotchas to note: * it only scans the paths `src/main` and `src/test` (see [docs](http://scala-tools.org/mvnsites/maven-scala-plugin/usage_cc.html)), so it will only work @@ -157,9 +168,9 @@ The debian package can then be found under assembly/target. We added the short c Running only Java 8 tests and nothing else. mvn install -DskipTests -Pjava8-tests - -Java 8 tests are run when `-Pjava8-tests` profile is enabled, they will run in spite of `-DskipTests`. -For these tests to run your system must have a JDK 8 installation. + +Java 8 tests are run when `-Pjava8-tests` profile is enabled, they will run in spite of `-DskipTests`. +For these tests to run your system must have a JDK 8 installation. If you have JDK 8 installed but it is not the system default, you can set JAVA_HOME to point to JDK 8 before running the tests. # Building for PySpark on YARN @@ -171,7 +182,7 @@ then ship it over to the cluster. We are investigating the exact cause for this. # Packaging without Hadoop Dependencies for YARN -The assembly jar produced by `mvn package` will, by default, include all of Spark's dependencies, including Hadoop and some of its ecosystem projects. On YARN deployments, this causes multiple versions of these to appear on executor classpaths: the version packaged in the Spark assembly and the version on each node, included with yarn.application.classpath. The `hadoop-provided` profile builds the assembly without including Hadoop-ecosystem projects, like ZooKeeper and Hadoop itself. +The assembly jar produced by `mvn package` will, by default, include all of Spark's dependencies, including Hadoop and some of its ecosystem projects. On YARN deployments, this causes multiple versions of these to appear on executor classpaths: the version packaged in the Spark assembly and the version on each node, included with yarn.application.classpath. The `hadoop-provided` profile builds the assembly without including Hadoop-ecosystem projects, like ZooKeeper and Hadoop itself. # Building with SBT @@ -182,22 +193,22 @@ compilation. More advanced developers may wish to use SBT. The SBT build is derived from the Maven POM files, and so the same Maven profiles and variables can be set to control the SBT build. For example: - sbt/sbt -Pyarn -Phadoop-2.3 assembly + build/sbt -Pyarn -Phadoop-2.3 assembly # Testing with SBT -Some of the tests require Spark to be packaged first, so always run `sbt/sbt assembly` the first time. The following is an example of a correct (build, test) sequence: +Some of the tests require Spark to be packaged first, so always run `build/sbt assembly` the first time. The following is an example of a correct (build, test) sequence: - sbt/sbt -Pyarn -Phadoop-2.3 -Phive -Phive-thriftserver assembly - sbt/sbt -Pyarn -Phadoop-2.3 -Phive -Phive-thriftserver test + build/sbt -Pyarn -Phadoop-2.3 -Phive -Phive-thriftserver assembly + build/sbt -Pyarn -Phadoop-2.3 -Phive -Phive-thriftserver test To run only a specific test suite as follows: - sbt/sbt -Pyarn -Phadoop-2.3 -Phive -Phive-thriftserver "test-only org.apache.spark.repl.ReplSuite" + build/sbt -Pyarn -Phadoop-2.3 -Phive -Phive-thriftserver "test-only org.apache.spark.repl.ReplSuite" To run test suites of a specific sub project as follows: - sbt/sbt -Pyarn -Phadoop-2.3 -Phive -Phive-thriftserver core/test + build/sbt -Pyarn -Phadoop-2.3 -Phive -Phive-thriftserver core/test # Speeding up Compilation with Zinc @@ -206,3 +217,9 @@ compiler. When run locally as a background process, it speeds up builds of Scala like Spark. Developers who regularly recompile Spark with Maven will be the most interested in Zinc. The project site gives instructions for building and running `zinc`; OS X users can install it using `brew install zinc`. + +If using the `build/mvn` package `zinc` will automatically be downloaded and leveraged for all +builds. This process will auto-start after the first time `build/mvn` is called and bind to port +3030 unless the `ZINC_PORT` environment variable is set. The `zinc` process can subsequently be +shut down at any time by running `build/zinc-/bin/zinc -shutdown` and will automatically +restart whenever `build/mvn` is called. diff --git a/docs/hadoop-third-party-distributions.md b/docs/hadoop-third-party-distributions.md index dd73e9dc54..87dcc58feb 100644 --- a/docs/hadoop-third-party-distributions.md +++ b/docs/hadoop-third-party-distributions.md @@ -18,7 +18,7 @@ see the guide on [building with maven](building-spark.html#specifying-the-hadoop The table below lists the corresponding `hadoop.version` code for each CDH/HDP release. Note that some Hadoop releases are binary compatible across client versions. This means the pre-built Spark -distribution may "just work" without you needing to compile. That said, we recommend compiling with +distribution may "just work" without you needing to compile. That said, we recommend compiling with the _exact_ Hadoop version you are running to avoid any compatibility errors. @@ -50,7 +50,7 @@ the _exact_ Hadoop version you are running to avoid any compatibility errors. In SBT, the equivalent can be achieved by setting the the `hadoop.version` property: - sbt/sbt -Dhadoop.version=1.0.4 assembly + build/sbt -Dhadoop.version=1.0.4 assembly # Linking Applications to the Hadoop Version @@ -98,11 +98,11 @@ Spark can run in a variety of deployment modes: * Using dedicated set of Spark nodes in your cluster. These nodes should be co-located with your Hadoop installation. -* Running on the same nodes as an existing Hadoop installation, with a fixed amount memory and +* Running on the same nodes as an existing Hadoop installation, with a fixed amount memory and cores dedicated to Spark on each node. * Run Spark alongside Hadoop using a cluster resource manager, such as YARN or Mesos. -These options are identical for those using CDH and HDP. +These options are identical for those using CDH and HDP. # Inheriting Cluster Configuration @@ -116,5 +116,5 @@ The location of these configuration files varies across CDH and HDP versions, bu a common location is inside of `/etc/hadoop/conf`. Some tools, such as Cloudera Manager, create configurations on-the-fly, but offer a mechanisms to download copies of them. -To make these files visible to Spark, set `HADOOP_CONF_DIR` in `$SPARK_HOME/spark-env.sh` +To make these files visible to Spark, set `HADOOP_CONF_DIR` in `$SPARK_HOME/spark-env.sh` to a location containing the configuration files. diff --git a/extras/java8-tests/README.md b/extras/java8-tests/README.md index e95b73ac77..dc9e87f2ee 100644 --- a/extras/java8-tests/README.md +++ b/extras/java8-tests/README.md @@ -8,7 +8,7 @@ to your Java location. The set-up depends a bit on the build system: `-java-home` to the sbt launch script. If a Java 8 JDK is detected sbt will automatically include the Java 8 test project. - `$ JAVA_HOME=/opt/jdk1.8.0/ sbt/sbt clean "test-only org.apache.spark.Java8APISuite"` + `$ JAVA_HOME=/opt/jdk1.8.0/ build/sbt clean "test-only org.apache.spark.Java8APISuite"` * For Maven users, @@ -19,6 +19,6 @@ to your Java location. The set-up depends a bit on the build system: `$ JAVA_HOME=/opt/jdk1.8.0/ mvn clean install -DskipTests` `$ JAVA_HOME=/opt/jdk1.8.0/ mvn test -Pjava8-tests -DwildcardSuites=org.apache.spark.Java8APISuite` - Note that the above command can only be run from project root directory since this module - depends on core and the test-jars of core and streaming. This means an install step is + Note that the above command can only be run from project root directory since this module + depends on core and the test-jars of core and streaming. This means an install step is required to make the test dependencies visible to the Java 8 sub-project. diff --git a/python/pyspark/sql.py b/python/pyspark/sql.py index 9807a84a66..0e8b398fc6 100644 --- a/python/pyspark/sql.py +++ b/python/pyspark/sql.py @@ -1671,7 +1671,7 @@ class HiveContext(SQLContext): except Py4JError as e: raise Exception("You must build Spark with Hive. " "Export 'SPARK_HIVE=true' and run " - "sbt/sbt assembly", e) + "build/sbt assembly", e) def _get_hive_ctx(self): return self._jvm.HiveContext(self._jsc.sc()) diff --git a/sbt/sbt b/sbt/sbt index 0a251d97db..6f3e5e08ed 100755 --- a/sbt/sbt +++ b/sbt/sbt @@ -1,111 +1,12 @@ -#!/usr/bin/env bash +#!/bin/bash -# When creating new tests for Spark SQL Hive, the HADOOP_CLASSPATH must contain the hive jars so -# that we can run Hive to generate the golden answer. This is not required for normal development -# or testing. -for i in "$HIVE_HOME"/lib/* -do HADOOP_CLASSPATH="$HADOOP_CLASSPATH:$i" -done -export HADOOP_CLASSPATH +# Determine the current working directory +_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )" -realpath () { -( - TARGET_FILE="$1" +echo "NOTE: The sbt/sbt script has been relocated to build/sbt." >&2 +echo " Please update references to point to the new location." >&2 +echo "" >&2 +echo " Invoking 'build/sbt $@' now ..." >&2 +echo "" >&2 - cd "$(dirname "$TARGET_FILE")" - TARGET_FILE="$(basename "$TARGET_FILE")" - - COUNT=0 - while [ -L "$TARGET_FILE" -a $COUNT -lt 100 ] - do - TARGET_FILE="$(readlink "$TARGET_FILE")" - cd $(dirname "$TARGET_FILE") - TARGET_FILE="$(basename $TARGET_FILE)" - COUNT=$(($COUNT + 1)) - done - - echo "$(pwd -P)/"$TARGET_FILE"" -) -} - -. "$(dirname "$(realpath "$0")")"/sbt-launch-lib.bash - - -declare -r noshare_opts="-Dsbt.global.base=project/.sbtboot -Dsbt.boot.directory=project/.boot -Dsbt.ivy.home=project/.ivy" -declare -r sbt_opts_file=".sbtopts" -declare -r etc_sbt_opts_file="/etc/sbt/sbtopts" - -usage() { - cat < path to global settings/plugins directory (default: ~/.sbt) - -sbt-boot path to shared boot directory (default: ~/.sbt/boot in 0.11 series) - -ivy path to local Ivy repository (default: ~/.ivy2) - -mem set memory options (default: $sbt_mem, which is $(get_mem_opts $sbt_mem)) - -no-share use all local caches; no sharing - -no-global uses global caches, but does not use global ~/.sbt directory. - -jvm-debug Turn on JVM debugging, open at the given port. - -batch Disable interactive mode - - # sbt version (default: from project/build.properties if present, else latest release) - -sbt-version use the specified version of sbt - -sbt-jar use the specified jar as the sbt launcher - -sbt-rc use an RC version of sbt - -sbt-snapshot use a snapshot version of sbt - - # java version (default: java from PATH, currently $(java -version 2>&1 | grep version)) - -java-home alternate JAVA_HOME - - # jvm options and output control - JAVA_OPTS environment variable, if unset uses "$java_opts" - SBT_OPTS environment variable, if unset uses "$default_sbt_opts" - .sbtopts if this file exists in the current directory, it is - prepended to the runner args - /etc/sbt/sbtopts if this file exists, it is prepended to the runner args - -Dkey=val pass -Dkey=val directly to the java runtime - -J-X pass option -X directly to the java runtime - (-J is stripped) - -S-X add -X to sbt's scalacOptions (-S is stripped) - -PmavenProfiles Enable a maven profile for the build. - -In the case of duplicated or conflicting options, the order above -shows precedence: JAVA_OPTS lowest, command line options highest. -EOM -} - -process_my_args () { - while [[ $# -gt 0 ]]; do - case "$1" in - -no-colors) addJava "-Dsbt.log.noformat=true" && shift ;; - -no-share) addJava "$noshare_opts" && shift ;; - -no-global) addJava "-Dsbt.global.base=$(pwd)/project/.sbtboot" && shift ;; - -sbt-boot) require_arg path "$1" "$2" && addJava "-Dsbt.boot.directory=$2" && shift 2 ;; - -sbt-dir) require_arg path "$1" "$2" && addJava "-Dsbt.global.base=$2" && shift 2 ;; - -debug-inc) addJava "-Dxsbt.inc.debug=true" && shift ;; - -batch) exec