spark-instrumented-optimizer/dev/run-pip-tests

137 lines
4.7 KiB
Plaintext
Raw Normal View History

[SPARK-1267][SPARK-18129] Allow PySpark to be pip installed ## What changes were proposed in this pull request? This PR aims to provide a pip installable PySpark package. This does a bunch of work to copy the jars over and package them with the Python code (to prevent challenges from trying to use different versions of the Python code with different versions of the JAR). It does not currently publish to PyPI but that is the natural follow up (SPARK-18129). Done: - pip installable on conda [manual tested] - setup.py installed on a non-pip managed system (RHEL) with YARN [manual tested] - Automated testing of this (virtualenv) - packaging and signing with release-build* Possible follow up work: - release-build update to publish to PyPI (SPARK-18128) - figure out who owns the pyspark package name on prod PyPI (is it someone with in the project or should we ask PyPI or should we choose a different name to publish with like ApachePySpark?) - Windows support and or testing ( SPARK-18136 ) - investigate details of wheel caching and see if we can avoid cleaning the wheel cache during our test - consider how we want to number our dev/snapshot versions Explicitly out of scope: - Using pip installed PySpark to start a standalone cluster - Using pip installed PySpark for non-Python Spark programs *I've done some work to test release-build locally but as a non-committer I've just done local testing. ## How was this patch tested? Automated testing with virtualenv, manual testing with conda, a system wide install, and YARN integration. release-build changes tested locally as a non-committer (no testing of upload artifacts to Apache staging websites) Author: Holden Karau <holden@us.ibm.com> Author: Juliet Hougland <juliet@cloudera.com> Author: Juliet Hougland <not@myemail.com> Closes #15659 from holdenk/SPARK-1267-pip-install-pyspark.
2016-11-16 17:22:15 -05:00
#!/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.
#
# Stop on error
set -e
# Set nullglob for when we are checking existence based on globs
shopt -s nullglob
FWDIR="$(cd "$(dirname "$0")"/..; pwd)"
cd "$FWDIR"
echo "Constructing virtual env for testing"
[SPARK-1267][SPARK-18129] Allow PySpark to be pip installed ## What changes were proposed in this pull request? This PR aims to provide a pip installable PySpark package. This does a bunch of work to copy the jars over and package them with the Python code (to prevent challenges from trying to use different versions of the Python code with different versions of the JAR). It does not currently publish to PyPI but that is the natural follow up (SPARK-18129). Done: - pip installable on conda [manual tested] - setup.py installed on a non-pip managed system (RHEL) with YARN [manual tested] - Automated testing of this (virtualenv) - packaging and signing with release-build* Possible follow up work: - release-build update to publish to PyPI (SPARK-18128) - figure out who owns the pyspark package name on prod PyPI (is it someone with in the project or should we ask PyPI or should we choose a different name to publish with like ApachePySpark?) - Windows support and or testing ( SPARK-18136 ) - investigate details of wheel caching and see if we can avoid cleaning the wheel cache during our test - consider how we want to number our dev/snapshot versions Explicitly out of scope: - Using pip installed PySpark to start a standalone cluster - Using pip installed PySpark for non-Python Spark programs *I've done some work to test release-build locally but as a non-committer I've just done local testing. ## How was this patch tested? Automated testing with virtualenv, manual testing with conda, a system wide install, and YARN integration. release-build changes tested locally as a non-committer (no testing of upload artifacts to Apache staging websites) Author: Holden Karau <holden@us.ibm.com> Author: Juliet Hougland <juliet@cloudera.com> Author: Juliet Hougland <not@myemail.com> Closes #15659 from holdenk/SPARK-1267-pip-install-pyspark.
2016-11-16 17:22:15 -05:00
VIRTUALENV_BASE=$(mktemp -d)
# Clean up the virtual env environment used if we created one.
[SPARK-1267][SPARK-18129] Allow PySpark to be pip installed ## What changes were proposed in this pull request? This PR aims to provide a pip installable PySpark package. This does a bunch of work to copy the jars over and package them with the Python code (to prevent challenges from trying to use different versions of the Python code with different versions of the JAR). It does not currently publish to PyPI but that is the natural follow up (SPARK-18129). Done: - pip installable on conda [manual tested] - setup.py installed on a non-pip managed system (RHEL) with YARN [manual tested] - Automated testing of this (virtualenv) - packaging and signing with release-build* Possible follow up work: - release-build update to publish to PyPI (SPARK-18128) - figure out who owns the pyspark package name on prod PyPI (is it someone with in the project or should we ask PyPI or should we choose a different name to publish with like ApachePySpark?) - Windows support and or testing ( SPARK-18136 ) - investigate details of wheel caching and see if we can avoid cleaning the wheel cache during our test - consider how we want to number our dev/snapshot versions Explicitly out of scope: - Using pip installed PySpark to start a standalone cluster - Using pip installed PySpark for non-Python Spark programs *I've done some work to test release-build locally but as a non-committer I've just done local testing. ## How was this patch tested? Automated testing with virtualenv, manual testing with conda, a system wide install, and YARN integration. release-build changes tested locally as a non-committer (no testing of upload artifacts to Apache staging websites) Author: Holden Karau <holden@us.ibm.com> Author: Juliet Hougland <juliet@cloudera.com> Author: Juliet Hougland <not@myemail.com> Closes #15659 from holdenk/SPARK-1267-pip-install-pyspark.
2016-11-16 17:22:15 -05:00
function delete_virtualenv() {
echo "Cleaning up temporary directory - $VIRTUALENV_BASE"
rm -rf "$VIRTUALENV_BASE"
}
trap delete_virtualenv EXIT
PYTHON_EXECS=()
[SPARK-1267][SPARK-18129] Allow PySpark to be pip installed ## What changes were proposed in this pull request? This PR aims to provide a pip installable PySpark package. This does a bunch of work to copy the jars over and package them with the Python code (to prevent challenges from trying to use different versions of the Python code with different versions of the JAR). It does not currently publish to PyPI but that is the natural follow up (SPARK-18129). Done: - pip installable on conda [manual tested] - setup.py installed on a non-pip managed system (RHEL) with YARN [manual tested] - Automated testing of this (virtualenv) - packaging and signing with release-build* Possible follow up work: - release-build update to publish to PyPI (SPARK-18128) - figure out who owns the pyspark package name on prod PyPI (is it someone with in the project or should we ask PyPI or should we choose a different name to publish with like ApachePySpark?) - Windows support and or testing ( SPARK-18136 ) - investigate details of wheel caching and see if we can avoid cleaning the wheel cache during our test - consider how we want to number our dev/snapshot versions Explicitly out of scope: - Using pip installed PySpark to start a standalone cluster - Using pip installed PySpark for non-Python Spark programs *I've done some work to test release-build locally but as a non-committer I've just done local testing. ## How was this patch tested? Automated testing with virtualenv, manual testing with conda, a system wide install, and YARN integration. release-build changes tested locally as a non-committer (no testing of upload artifacts to Apache staging websites) Author: Holden Karau <holden@us.ibm.com> Author: Juliet Hougland <juliet@cloudera.com> Author: Juliet Hougland <not@myemail.com> Closes #15659 from holdenk/SPARK-1267-pip-install-pyspark.
2016-11-16 17:22:15 -05:00
# Some systems don't have pip or virtualenv - in those cases our tests won't work.
if hash virtualenv 2>/dev/null && [ ! -n "$USE_CONDA" ]; then
echo "virtualenv installed - using. Note if this is a conda virtual env you may wish to set USE_CONDA"
# Figure out which Python execs we should test pip installation with
if hash python2 2>/dev/null; then
# We do this since we are testing with virtualenv and the default virtual env python
# is in /usr/bin/python
PYTHON_EXECS+=('python2')
elif hash python 2>/dev/null; then
# If python2 isn't installed fallback to python if available
PYTHON_EXECS+=('python')
fi
if hash python3 2>/dev/null; then
PYTHON_EXECS+=('python3')
fi
elif hash conda 2>/dev/null; then
echo "Using conda virtual enviroments"
PYTHON_EXECS=('3.5')
USE_CONDA=1
else
echo "Missing virtualenv & conda, skipping pip installability tests"
[SPARK-1267][SPARK-18129] Allow PySpark to be pip installed ## What changes were proposed in this pull request? This PR aims to provide a pip installable PySpark package. This does a bunch of work to copy the jars over and package them with the Python code (to prevent challenges from trying to use different versions of the Python code with different versions of the JAR). It does not currently publish to PyPI but that is the natural follow up (SPARK-18129). Done: - pip installable on conda [manual tested] - setup.py installed on a non-pip managed system (RHEL) with YARN [manual tested] - Automated testing of this (virtualenv) - packaging and signing with release-build* Possible follow up work: - release-build update to publish to PyPI (SPARK-18128) - figure out who owns the pyspark package name on prod PyPI (is it someone with in the project or should we ask PyPI or should we choose a different name to publish with like ApachePySpark?) - Windows support and or testing ( SPARK-18136 ) - investigate details of wheel caching and see if we can avoid cleaning the wheel cache during our test - consider how we want to number our dev/snapshot versions Explicitly out of scope: - Using pip installed PySpark to start a standalone cluster - Using pip installed PySpark for non-Python Spark programs *I've done some work to test release-build locally but as a non-committer I've just done local testing. ## How was this patch tested? Automated testing with virtualenv, manual testing with conda, a system wide install, and YARN integration. release-build changes tested locally as a non-committer (no testing of upload artifacts to Apache staging websites) Author: Holden Karau <holden@us.ibm.com> Author: Juliet Hougland <juliet@cloudera.com> Author: Juliet Hougland <not@myemail.com> Closes #15659 from holdenk/SPARK-1267-pip-install-pyspark.
2016-11-16 17:22:15 -05:00
exit 0
fi
if ! hash pip 2>/dev/null; then
echo "Missing pip, skipping pip installability tests."
exit 0
fi
# Determine which version of PySpark we are building for archive name
PYSPARK_VERSION=$(python3 -c "exec(open('python/pyspark/version.py').read());print(__version__)")
[SPARK-1267][SPARK-18129] Allow PySpark to be pip installed ## What changes were proposed in this pull request? This PR aims to provide a pip installable PySpark package. This does a bunch of work to copy the jars over and package them with the Python code (to prevent challenges from trying to use different versions of the Python code with different versions of the JAR). It does not currently publish to PyPI but that is the natural follow up (SPARK-18129). Done: - pip installable on conda [manual tested] - setup.py installed on a non-pip managed system (RHEL) with YARN [manual tested] - Automated testing of this (virtualenv) - packaging and signing with release-build* Possible follow up work: - release-build update to publish to PyPI (SPARK-18128) - figure out who owns the pyspark package name on prod PyPI (is it someone with in the project or should we ask PyPI or should we choose a different name to publish with like ApachePySpark?) - Windows support and or testing ( SPARK-18136 ) - investigate details of wheel caching and see if we can avoid cleaning the wheel cache during our test - consider how we want to number our dev/snapshot versions Explicitly out of scope: - Using pip installed PySpark to start a standalone cluster - Using pip installed PySpark for non-Python Spark programs *I've done some work to test release-build locally but as a non-committer I've just done local testing. ## How was this patch tested? Automated testing with virtualenv, manual testing with conda, a system wide install, and YARN integration. release-build changes tested locally as a non-committer (no testing of upload artifacts to Apache staging websites) Author: Holden Karau <holden@us.ibm.com> Author: Juliet Hougland <juliet@cloudera.com> Author: Juliet Hougland <not@myemail.com> Closes #15659 from holdenk/SPARK-1267-pip-install-pyspark.
2016-11-16 17:22:15 -05:00
PYSPARK_DIST="$FWDIR/python/dist/pyspark-$PYSPARK_VERSION.tar.gz"
# The pip install options we use for all the pip commands
PIP_OPTIONS="--upgrade --no-cache-dir --force-reinstall "
# Test both regular user and edit/dev install modes.
PIP_COMMANDS=("pip install $PIP_OPTIONS $PYSPARK_DIST"
"pip install $PIP_OPTIONS -e python/")
for python in "${PYTHON_EXECS[@]}"; do
for install_command in "${PIP_COMMANDS[@]}"; do
echo "Testing pip installation with python $python"
# Create a temp directory for us to work in and save its name to a file for cleanup
echo "Using $VIRTUALENV_BASE for virtualenv"
VIRTUALENV_PATH="$VIRTUALENV_BASE"/$python
rm -rf "$VIRTUALENV_PATH"
if [ -n "$USE_CONDA" ]; then
conda create -y -p "$VIRTUALENV_PATH" python=$python numpy pandas pip setuptools
source activate "$VIRTUALENV_PATH"
else
mkdir -p "$VIRTUALENV_PATH"
virtualenv --python=$python "$VIRTUALENV_PATH"
source "$VIRTUALENV_PATH"/bin/activate
fi
# Upgrade pip & friends if using virutal env
if [ ! -n "USE_CONDA" ]; then
pip install --upgrade pip pypandoc wheel numpy
fi
[SPARK-1267][SPARK-18129] Allow PySpark to be pip installed ## What changes were proposed in this pull request? This PR aims to provide a pip installable PySpark package. This does a bunch of work to copy the jars over and package them with the Python code (to prevent challenges from trying to use different versions of the Python code with different versions of the JAR). It does not currently publish to PyPI but that is the natural follow up (SPARK-18129). Done: - pip installable on conda [manual tested] - setup.py installed on a non-pip managed system (RHEL) with YARN [manual tested] - Automated testing of this (virtualenv) - packaging and signing with release-build* Possible follow up work: - release-build update to publish to PyPI (SPARK-18128) - figure out who owns the pyspark package name on prod PyPI (is it someone with in the project or should we ask PyPI or should we choose a different name to publish with like ApachePySpark?) - Windows support and or testing ( SPARK-18136 ) - investigate details of wheel caching and see if we can avoid cleaning the wheel cache during our test - consider how we want to number our dev/snapshot versions Explicitly out of scope: - Using pip installed PySpark to start a standalone cluster - Using pip installed PySpark for non-Python Spark programs *I've done some work to test release-build locally but as a non-committer I've just done local testing. ## How was this patch tested? Automated testing with virtualenv, manual testing with conda, a system wide install, and YARN integration. release-build changes tested locally as a non-committer (no testing of upload artifacts to Apache staging websites) Author: Holden Karau <holden@us.ibm.com> Author: Juliet Hougland <juliet@cloudera.com> Author: Juliet Hougland <not@myemail.com> Closes #15659 from holdenk/SPARK-1267-pip-install-pyspark.
2016-11-16 17:22:15 -05:00
echo "Creating pip installable source dist"
cd "$FWDIR"/python
# Delete the egg info file if it exists, this can cache the setup file.
rm -rf pyspark.egg-info || echo "No existing egg info file, skipping deletion"
python setup.py sdist
[SPARK-1267][SPARK-18129] Allow PySpark to be pip installed ## What changes were proposed in this pull request? This PR aims to provide a pip installable PySpark package. This does a bunch of work to copy the jars over and package them with the Python code (to prevent challenges from trying to use different versions of the Python code with different versions of the JAR). It does not currently publish to PyPI but that is the natural follow up (SPARK-18129). Done: - pip installable on conda [manual tested] - setup.py installed on a non-pip managed system (RHEL) with YARN [manual tested] - Automated testing of this (virtualenv) - packaging and signing with release-build* Possible follow up work: - release-build update to publish to PyPI (SPARK-18128) - figure out who owns the pyspark package name on prod PyPI (is it someone with in the project or should we ask PyPI or should we choose a different name to publish with like ApachePySpark?) - Windows support and or testing ( SPARK-18136 ) - investigate details of wheel caching and see if we can avoid cleaning the wheel cache during our test - consider how we want to number our dev/snapshot versions Explicitly out of scope: - Using pip installed PySpark to start a standalone cluster - Using pip installed PySpark for non-Python Spark programs *I've done some work to test release-build locally but as a non-committer I've just done local testing. ## How was this patch tested? Automated testing with virtualenv, manual testing with conda, a system wide install, and YARN integration. release-build changes tested locally as a non-committer (no testing of upload artifacts to Apache staging websites) Author: Holden Karau <holden@us.ibm.com> Author: Juliet Hougland <juliet@cloudera.com> Author: Juliet Hougland <not@myemail.com> Closes #15659 from holdenk/SPARK-1267-pip-install-pyspark.
2016-11-16 17:22:15 -05:00
echo "Installing dist into virtual env"
cd dist
# Verify that the dist directory only contains one thing to install
sdists=(*.tar.gz)
if [ ${#sdists[@]} -ne 1 ]; then
echo "Unexpected number of targets found in dist directory - please cleanup existing sdists first."
exit -1
fi
# Do the actual installation
cd "$FWDIR"
$install_command
cd /
echo "Run basic sanity check on pip installed version with spark-submit"
spark-submit "$FWDIR"/dev/pip-sanity-check.py
echo "Run basic sanity check with import based"
python "$FWDIR"/dev/pip-sanity-check.py
echo "Run the tests for context.py"
python "$FWDIR"/python/pyspark/context.py
cd "$FWDIR"
# conda / virtualenv enviroments need to be deactivated differently
if [ -n "$USE_CONDA" ]; then
source deactivate
else
deactivate
fi
[SPARK-1267][SPARK-18129] Allow PySpark to be pip installed ## What changes were proposed in this pull request? This PR aims to provide a pip installable PySpark package. This does a bunch of work to copy the jars over and package them with the Python code (to prevent challenges from trying to use different versions of the Python code with different versions of the JAR). It does not currently publish to PyPI but that is the natural follow up (SPARK-18129). Done: - pip installable on conda [manual tested] - setup.py installed on a non-pip managed system (RHEL) with YARN [manual tested] - Automated testing of this (virtualenv) - packaging and signing with release-build* Possible follow up work: - release-build update to publish to PyPI (SPARK-18128) - figure out who owns the pyspark package name on prod PyPI (is it someone with in the project or should we ask PyPI or should we choose a different name to publish with like ApachePySpark?) - Windows support and or testing ( SPARK-18136 ) - investigate details of wheel caching and see if we can avoid cleaning the wheel cache during our test - consider how we want to number our dev/snapshot versions Explicitly out of scope: - Using pip installed PySpark to start a standalone cluster - Using pip installed PySpark for non-Python Spark programs *I've done some work to test release-build locally but as a non-committer I've just done local testing. ## How was this patch tested? Automated testing with virtualenv, manual testing with conda, a system wide install, and YARN integration. release-build changes tested locally as a non-committer (no testing of upload artifacts to Apache staging websites) Author: Holden Karau <holden@us.ibm.com> Author: Juliet Hougland <juliet@cloudera.com> Author: Juliet Hougland <not@myemail.com> Closes #15659 from holdenk/SPARK-1267-pip-install-pyspark.
2016-11-16 17:22:15 -05:00
done
done
exit 0