source: other-projects/hathitrust/solr-extracted-features/trunk/_RUN.bash@ 30929

#!/bin/bash

# To work, the follow bash variables need to have been set:
#
#  json_filelist input_dir output_dir
#
# Typically done through running a wrapper script, such as:
#
#  RUN-PD-CLUSTER.bash

if [ "x$json_filelist" = "x" ] ; then
    echo "_RUN.bash: Failed to set 'json_filelist'" 1>&2
    exit
fi

if [ "x$input_dir" = "x" ] ; then
    echo "_RUN.bash: Failed to set 'input_dir'" 1>&2
    exit
fi

if [ "x$output_dir" = "x" ] ; then
    echo "_RUN.bash: Failed to set 'output_dir'" 1>&2
    exit
fi

echo
echo "****"
echo "* Checking for Spark and Hadoop daemons"
echo "****"
jps | sed 's/^/* /g'
echo "****"
echo "* Done"
echo "****"
echo

self_contained_jar=target/htrc-ef-ingest-0.9-jar-with-dependencies.jar
base_cmd="spark-submit --class org.hathitrust.PrepareForIngest $master_opt $self_contained_jar"

cmd="$base_cmd --json-filelist=\"$json_filelist\" $input_dir $output_dir $*"

echo "****"
echo "* Lauching:"
echo "*   $cmd"
echo "****"
echo "* Monitor progress through:"
echo "*   http://10.10.0.52:8080/"
echo "****"
echo

$cmd

#    spark-submit --class org.hathitrust.PrepareForIngest --master local[4] target/htrc-ef-ingest-0.9-jar-with-dependencies.jar --json-filelist=pd-file-listing-step10000.txt pd-ef-json-files pd-solr-json-files $* 

# spark-submit --class org.hathitrust.PrepareForIngest --master local[4] target\htrc-ef-ingest-0.9-jar-with-dependencies.jar --json-filelist=pd-file-listing-step1000.txt json-files solr-files $*