spark-instrumented-optimizer/dev/test-dependencies.sh
Josh Rosen 3ab0138b0f [SPARK-12734][BUILD] Fix Netty exclusion and use Maven Enforcer to prevent future bugs
Netty classes are published under multiple artifacts with different names, so our build needs to exclude the `io.netty:netty` and `org.jboss.netty:netty` versions of the Netty artifact. However, our existing exclusions were incomplete, leading to situations where duplicate Netty classes would wind up on the classpath and cause compile errors (or worse).

This patch fixes the exclusion issue by adding more exclusions and uses Maven Enforcer's [banned dependencies](https://maven.apache.org/enforcer/enforcer-rules/bannedDependencies.html) rule to prevent these classes from accidentally being reintroduced. I also updated `dev/test-dependencies.sh` to run `mvn validate` so that the enforcer rules can run as part of pull request builds.

/cc rxin srowen pwendell. I'd like to backport at least the exclusion portion of this fix to `branch-1.5` in order to fix the documentation publishing job, which fails nondeterministically due to incompatible versions of Netty classes taking precedence on the compile-time classpath.

Author: Josh Rosen <rosenville@gmail.com>
Author: Josh Rosen <joshrosen@databricks.com>

Closes #10672 from JoshRosen/enforce-netty-exclusions.
2016-01-10 19:59:01 -08:00

112 lines
3.7 KiB
Bash
Executable file

#!/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.
#
set -e
FWDIR="$(cd "`dirname $0`"/..; pwd)"
cd "$FWDIR"
# Explicitly set locale in order to make `sort` output consistent across machines.
# See https://stackoverflow.com/questions/28881 for more details.
export LC_ALL=C
# TODO: This would be much nicer to do in SBT, once SBT supports Maven-style resolution.
# NOTE: These should match those in the release publishing script
HADOOP2_MODULE_PROFILES="-Phive-thriftserver -Pyarn -Phive"
MVN="build/mvn --force"
HADOOP_PROFILES=(
hadoop-2.2
hadoop-2.3
hadoop-2.4
hadoop-2.6
)
# We'll switch the version to a temp. one, publish POMs using that new version, then switch back to
# the old version. We need to do this because the `dependency:build-classpath` task needs to
# resolve Spark's internal submodule dependencies.
# From http://stackoverflow.com/a/26514030
set +e
OLD_VERSION=$($MVN -q \
-Dexec.executable="echo" \
-Dexec.args='${project.version}' \
--non-recursive \
org.codehaus.mojo:exec-maven-plugin:1.3.1:exec)
if [ $? != 0 ]; then
echo -e "Error while getting version string from Maven:\n$OLD_VERSION"
exit 1
fi
set -e
TEMP_VERSION="spark-$(python -S -c "import random; print(random.randrange(100000, 999999))")"
function reset_version {
# Delete the temporary POMs that we wrote to the local Maven repo:
find "$HOME/.m2/" | grep "$TEMP_VERSION" | xargs rm -rf
# Restore the original version number:
$MVN -q versions:set -DnewVersion=$OLD_VERSION -DgenerateBackupPoms=false > /dev/null
}
trap reset_version EXIT
$MVN -q versions:set -DnewVersion=$TEMP_VERSION -DgenerateBackupPoms=false > /dev/null
# Generate manifests for each Hadoop profile:
for HADOOP_PROFILE in "${HADOOP_PROFILES[@]}"; do
echo "Performing Maven install for $HADOOP_PROFILE"
$MVN $HADOOP2_MODULE_PROFILES -P$HADOOP_PROFILE jar:jar jar:test-jar install:install -q
echo "Performing Maven validate for $HADOOP_PROFILE"
$MVN $HADOOP2_MODULE_PROFILES -P$HADOOP_PROFILE validate -q
echo "Generating dependency manifest for $HADOOP_PROFILE"
mkdir -p dev/pr-deps
$MVN $HADOOP2_MODULE_PROFILES -P$HADOOP_PROFILE dependency:build-classpath -pl assembly \
| grep "Building Spark Project Assembly" -A 5 \
| tail -n 1 | tr ":" "\n" | rev | cut -d "/" -f 1 | rev | sort \
| grep -v spark > dev/pr-deps/spark-deps-$HADOOP_PROFILE
done
if [[ $@ == **replace-manifest** ]]; then
echo "Replacing manifests and creating new files at dev/deps"
rm -rf dev/deps
mv dev/pr-deps dev/deps
exit 0
fi
for HADOOP_PROFILE in "${HADOOP_PROFILES[@]}"; do
set +e
dep_diff="$(
git diff \
--no-index \
dev/deps/spark-deps-$HADOOP_PROFILE \
dev/pr-deps/spark-deps-$HADOOP_PROFILE \
)"
set -e
if [ "$dep_diff" != "" ]; then
echo "Spark's published dependencies DO NOT MATCH the manifest file (dev/spark-deps)."
echo "To update the manifest file, run './dev/test-dependencies.sh --replace-manifest'."
echo "$dep_diff"
rm -rf dev/pr-deps
exit 1
fi
done
exit 0