From e9ea7e5dc0cd2c3456112ad46c754571ac6e555b Mon Sep 17 00:00:00 2001
From: jerryshao <sshao@hortonworks.com>
Date: Wed, 9 May 2018 13:41:41 +0800
Subject: [PATCH] Improve the docker building script to avoid copying
 everything under examples to docker image

Change-Id: Id4bad2b85a9038fddcfc007c44376be805b90bd1
---
 bin/docker-image-tool.sh                      | 26 +++++++++++++++++++
 .../src/main/dockerfiles/spark/Dockerfile     |  3 ++-
 2 files changed, 28 insertions(+), 1 deletion(-)

diff --git a/bin/docker-image-tool.sh b/bin/docker-image-tool.sh
index f090240065bf1..7ded77426340b 100755
--- a/bin/docker-image-tool.sh
+++ b/bin/docker-image-tool.sh
@@ -44,15 +44,37 @@ function image_ref {
 function build {
   local BUILD_ARGS
   local IMG_PATH
+  local TMPFOLDER
 
   if [ ! -f "$SPARK_HOME/RELEASE" ]; then
     # Set image build arguments accordingly if this is a source repo and not a distribution archive.
+    local JARS="${SPARK_HOME}/assembly/target/scala-${SPARK_SCALA_VERSION}/jars"
+    TMPFOLDER=`mktemp -q -d examples.XXXXXX`
+    if [ $? -ne 0 ]; then
+      ehco "Cannot create temp folder, exiting..."
+      exit 1
+    fi
+
+    mkdir -p "${TMPFOLDER}/jars"
+    cp "${SPARK_HOME}"/examples/target/scala*/jars/* "${TMPFOLDER}/jars"
+    for f in "${TMPFOLDER}"/jars/*; do
+      name=$(basename "$f")
+      if [ -f "${JARS}/${name}" ]; then
+        rm "${TMPFOLDER}/jars/${name}"
+      fi
+    done
+
+    mkdir -p "${TMPFOLDER}/src/main"
+    cp -r "${SPARK_HOME}/examples/src/main" "${TMPFOLDER}/src"
+
     IMG_PATH=resource-managers/kubernetes/docker/src/main/dockerfiles
     BUILD_ARGS=(
       --build-arg
       img_path=$IMG_PATH
       --build-arg
       spark_jars=assembly/target/scala-$SPARK_SCALA_VERSION/jars
+      --build-arg
+      spark_examples=$TMPFOLDER
     )
   else
     # Not passed as an argument to docker, but used to validate the Spark directory.
@@ -69,6 +91,10 @@ function build {
   docker build "${BUILD_ARGS[@]}" \
     -t $(image_ref spark) \
     -f "$DOCKERFILE" .
+
+  if [ -d "${TMPFOLDER}" ]; then
+    rm -fr "${TMPFOLDER}"
+  fi
 }
 
 function push {
diff --git a/resource-managers/kubernetes/docker/src/main/dockerfiles/spark/Dockerfile b/resource-managers/kubernetes/docker/src/main/dockerfiles/spark/Dockerfile
index 9badf8556afc3..198f14f2955d1 100644
--- a/resource-managers/kubernetes/docker/src/main/dockerfiles/spark/Dockerfile
+++ b/resource-managers/kubernetes/docker/src/main/dockerfiles/spark/Dockerfile
@@ -19,6 +19,7 @@ FROM openjdk:8-alpine
 
 ARG spark_jars=jars
 ARG img_path=kubernetes/dockerfiles
+ARG spark_examples=examples
 
 # Before building the docker image, first build and make a Spark distribution following
 # the instructions in http://spark.apache.org/docs/latest/building-spark.html.
@@ -41,7 +42,7 @@ COPY ${spark_jars} /opt/spark/jars
 COPY bin /opt/spark/bin
 COPY sbin /opt/spark/sbin
 COPY ${img_path}/spark/entrypoint.sh /opt/
-COPY examples /opt/spark/examples
+COPY ${spark_examples} /opt/spark/examples
 COPY data /opt/spark/data
 
 ENV SPARK_HOME /opt/spark