syntaxnet_v1.def

BootStrap: debootstrap
OSVersion: xenial
#MirrorURL: http://ftp.fau.de/ubuntu/
# Use the following URL in the US:
MirrorURL: http://us.archive.ubuntu.com/ubuntu/


%runscript
    cd /opt/models/syntaxnet
    sh /opt/multilingualpipeline/run.sh "$@"

%post
    /bin/bash
    echo "Hello from inside the container"
    sed -i 's/$/ universe/' /etc/apt/sources.list
    apt-get update
    apt-get upgrade -y
    apt-get install -y --force-yes git wget nano autoconf automake bzip2 gfortran  libatlas3-base liblapack-dev libopenblas-dev libtool python-dev unzip software-properties-common python-pip subversion curl ruby ruby-dev openjdk-8-jdk

    #Install Bazel
    echo "deb [arch=amd64] http://storage.googleapis.com/bazel-apt stable jdk1.8" | tee /etc/apt/sources.list.d/bazel.list
    curl https://bazel.build/bazel-release.pub.gpg | apt-key add -
    # Use -y
    apt-get update && apt-get install -y bazel
    #apt-get upgrade bazel
    apt-get install -y swig
    pip install --upgrade pip
    pip install -U protobuf==3.3.0
    pip install mock
    pip install asciitree
    pip install numpy
    apt-get install -y graphviz libgraphviz-dev
    pip install pygraphviz --install-option="--include-path=/usr/include/graphviz" --install-option="--library-path=/usr/lib/graphviz/"

    cd /opt
    git clone --recursive https://github.com/tensorflow/models.git
    cd models
    git checkout abeb03563717a644d9f03c8eb19ab96d11aa8012
    cd syntaxnet/tensorflow
    git checkout a7d6015d
    printf '/usr/bin/python\n-march=native\nn\nn\nn\nn\n/usr/local/lib/python2.7/dist-packages\nn\nn\nn' | ./configure
    cd ..
    bazel test ...

    mv /root/.cache /opt/.cache
    chmod -R a+rx /opt/.cache
    cd /opt/.cache/bazel/_bazel_root/
    find . -lname '/root/.cache*' -printf '%l\0%p\0' | sed -z 's|^/root|/opt|;n' | xargs -r0n2 ln -sfT
    cd /opt/models/syntaxnet/
    find . -lname '/root/.cache*' -printf '%l\0%p\0' | sed -z 's|^/root|/opt|;n' | xargs -r0n2 ln -sfT

    mkdir /tmp/syntaxnet_pkg
    bazel-bin/dragnn/tools/build_pip_package --output-dir=/tmp/syntaxnet_pkg
    # The following line requires a --cleanenv when I run it with "shell" Not sure yet how that will work during bootstrapping.
    pip install /tmp/syntaxnet_pkg/syntaxnet-0.2-cp27-cp27mu-linux_x86_64.whl
    rm -rf /tmp/syntaxnet_pkg

    cd /opt/models/syntaxnet
    chmod a+x syntaxnet/demo.sh

    cd /opt/models/syntaxnet/syntaxnet/models/
    mkdir other_language_models
    cd other_language_models
    wget http://download.tensorflow.org/models/parsey_universal/German.zip http://download.tensorflow.org/models/parsey_universal/Portuguese.zip http://download.tensorflow.org/models/parsey_universal/Russian-SynTagRus.zip http://download.tensorflow.org/models/parsey_universal/Norwegian.zip http://download.tensorflow.org/models/parsey_universal/Danish.zip http://download.tensorflow.org/models/parsey_universal/Swedish.zip http://download.tensorflow.org/models/parsey_universal/Polish.zip
    unzip '*.zip'
    chmod -R 755 .

    #Install CSTLemmatizer
    mkdir /opt/Lemmatizer/ && cd /opt/Lemmatizer/
    wget https://raw.githubusercontent.com/kuhumcst/cstlemma/master/doc/makecstlemma.bash
    sh makecstlemma.bash
    cd cstlemma/src
    #Download Rules for lemmatizing
    mkdir Rules && cd Rules
    mkdir Danish German Polish Portuguese Russian
    cd German && wget https://cst.dk/download/cstlemma/german/flexrules
    cd ../Danish && wget https://cst.dk/download/cstlemma/danish/dict https://cst.dk/download/cstlemma/danish/1/flexrules
    cd ../Polish && wget https://cst.dk/download/cstlemma/polish/dict https://cst.dk/download/cstlemma/polish/flexrules
    cd ../Portuguese && wget https://cst.dk/download/cstlemma/portuguese/dict https://cst.dk/download/cstlemma/portuguese/flexrules-supplement-with-dict
    cd ../Russian && wget https://cst.dk/download/cstlemma/russian/dict https://cst.dk/download/cstlemma/russian/flexrules2

    #Install Sentence Splitter
    gem install pragmatic_segmenter
    locale-gen --purge en_US.UTF-8

    #Install Treetagger
    mkdir /opt/Treetagger && cd /opt/Treetagger
    wget http://www.cis.uni-muenchen.de/~schmid/tools/TreeTagger/data/tree-tagger-linux-3.2.1.tar.gz http://www.cis.uni-muenchen.de/~schmid/tools/TreeTagger/data/tagger-scripts.tar.gz http://www.cis.uni-muenchen.de/~schmid/tools/TreeTagger/data/install-tagger.sh http://www.cis.uni-muenchen.de/~schmid/tools/TreeTagger/data/russian-par-linux-3.2-utf8.bin.gz
    sh install-tagger.sh
    cd lib
    wget http://person2.sol.lu.se/JohanFrid/webapps/treetagger-swe/model_tt_lower_lemmas #Download params for Swedish

    #Download Pipeline
    cd /opt
    git clone https://github.com/RedHenLab/multilingualpipeline.git
    cd multilingualpipeline
    mv make_sents.py /opt/Treetagger/
    mv tree-tagger-swedish /opt/Treetagger/cmd/