source: main/trunk/model-sites-dev/eurovision-lod/collect/eurovision/prepare/01-DOWNLOAD-ESC-LOD-DATA.sh@ 35904

#!/bin/bash

# Before the Errata was introduced, the SPARQL query could be run
# directly against the DBPedia endpoint:
#
##echo "Downloading SPARQL query: countries-in-esc-by-year.sparql"
##./DBPEDIA-LOD-SPARQL-QUERY.sh countries-in-esc-by-year.sparql

# Now a more complicated federated search is used.
# Because the DBPedia endpoint doesn't allow for queries that
# use SERIVCE to combine results from other endpoints, the
# query we issue is done on the local Greenstone Apache Jena endpoint,
# (which does have this facility on), where the Errata information is
# stored, but as part of the query run reaches out to the DBPedia endpoint.

logfile="01-DOWNLOAD--runtime-log-$$.txt"
echo "Running bash script: $0" > "$logfile"
echo "" >> "$logfile" 

if [ $# == 0 ] ; then
    sparql_query_list="local--countries-in-esc-by-year-in-1956--with-errata.sparql local--countries-in-esc-by-year-after-1956--with-errata.sparql"

    echo ""
    echo "Away to run the query:"
    echo ""
    echo "  $sparql_query_list"
    echo ""

    echo "This can take a while to run.  Worth considering running with 'nohup':"
    echo "  nohup $0 > nohup-01-download.out 2>&1 &"
    echo ""

    echo "Sleeping for 5 seconds as an opportunity to ^C the script and rerun it"
    echo "..."
    sleep 5
    echo "... continuing"
    
else
    sparql_query_list="$*"
fi

for f in $sparql_query_list ; do

    echo "Downloading JSON content via SPARQL query:" | tee -a "$logfile"
    echo " errata-lod/$f" | tee -a "$logfile"

    cd errata-lod \
    && date >> "../$logfile" \
    && ./LOCAL-LOD-QUERY.sh "$f"  \
    && date >> "../$logfile" \
    && cd ..

    echo "" >> "$logfile"
    
done

echo ""
echo "Timed logged output:"
echo "  $logfile"
echo ""

# cd errata-lod \
#    && date > runtime-log-$$.txt && ./LOCAL-LOD-QUERY.sh ./local--countries-in-esc-by-year-after-1956--with-errata.sparql  && date >> runtime-log-$$.txt