Docker image that builds parquet-tools and allows you to run parquet-tools

Dockerfile.parquet-tools

# Docker image that will allow one to run parquet-tools on any platform that supports docker.
#
# Build like this:
#    $ docker build -f Dockerfile -t jlinoff/parquet:latest .
#
# Run it like this in the directory that contains your parquet file.
#    $ docker run -it --rm --init -h jpt --name jpt -v $(pwd):/mnt jlinoff/parquet bash -i
#    $ docker run -it --rm --init -h jpt --name jpt -v $(pwd):/mnt jlinoff/parquet jpt --help
#    $ docker run -it --rm --init -h jpt --name jpt -v $(pwd):/mnt jlinoff/parquet jpt column-sizes test.parquet
#
# Note that jpt is just a wrapper around the parquet-tools program.
FROM openjdk:latest

# Initial setup.
RUN yum update -y && \
    yum group install -y "Development Tools" && \
    yum install -y tar tree wget curl git && \
    yum install -y boost boost-devel

# Setup the build area.
RUN mkdir -p /opt/parquet
WORKDIR /opt/parquet

# Build and install thrift.
# wget -nv http://archive.apache.org/dist/thrift/0.12.0/thrift-0.12.0.tar.gz
RUN curl -L http://archive.apache.org/dist/thrift/0.12.0/thrift-0.12.0.tar.gz --output /tmp/thrift.tar.gz && \
    tar xzf /tmp/thrift.tar.gz && \
    cd thrift-0.12.0 && \
    ./bootstrap.sh && \
    chmod +x ./configure && \
    ./configure --disable-libs && \
    make install && \
    rm -f /tmp/thrift.tar.gz && \
    cd -

# Install maven.
RUN curl -L https://apache.mirrors.lucidnetworks.net/maven/maven-3/3.6.3/binaries/apache-maven-3.6.3-bin.tar.gz --output /tmp/maven.tar.gz && \
    tar xzvf /tmp/maven.tar.gz && \
    rm -f /tmp/maven.tar.gz && \
    ls -l /opt/parquet/apache-maven-3.6.3/bin/mvn

# Update the path environment variable so that mvn can found.
ENV PATH="${PATH}:/opt/parquet/apache-maven-3.6.3/bin"

# Verify the installation of thrift and maven.
RUN ls -l /opt && \
    ls -l /opt/parquet && \
    mvn --version && \
    thrift --version && \
    java --version

# Build parquet tools.
# Skipping tests is not a great idea but they take a really
# long time to run. They can be re-enabled once this image
# is stabilized.
RUN git clone https://github.com/apache/parquet-mr.git && \
    cd parquet-mr && \
    LC_ALL=C mvn -pl :parquet-tools -am -Plocal -DskipTests clean install && \
    cd -

# Create the jpt (joe's parquet tool interface) script.
RUN echo '#!/usr/bin/env bash' > /usr/local/bin/jpt && \
    echo 'java -jar /opt/parquet/parquet-mr/parquet-tools/target/parquet-tools-1.12.0-SNAPSHOT.jar $*' >> /usr/local/bin/jpt && \
    chmod a+x  /usr/local/bin/jpt

# Set the default work directory.
WORKDIR /mnt