Skip to content

Commit

Permalink
update scripts to use warc-indexer 3.3.1
Browse files Browse the repository at this point in the history
  • Loading branch information
thomasegense committed Dec 23, 2023
1 parent ebbaf37 commit f9ca772
Show file tree
Hide file tree
Showing 6 changed files with 6 additions and 6 deletions.
2 changes: 1 addition & 1 deletion src/bundle/indexing/batch_warcs1_folder.bat
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
cd /D "%~dp0"

FOR /R warcs1 %%G IN (*.*) do java -Dfile.encoding=UTF-8 -Xmx2048M -Djava.io.tmpdir=tika_tmp -jar warc-indexer-3.2.0-SNAPSHOT-jar-with-dependencies.jar -c config3.conf -s "http://localhost:8983/solr/netarchivebuilder" "%%G"
FOR /R warcs1 %%G IN (*.*) do java -Dfile.encoding=UTF-8 -Xmx2048M -Djava.io.tmpdir=tika_tmp -jar warc-indexer-3.3.1-jar-with-dependencies.jar -c config3.conf -s "http://localhost:8983/solr/netarchivebuilder" "%%G"

2 changes: 1 addition & 1 deletion src/bundle/indexing/batch_warcs1_folder.sh
Original file line number Diff line number Diff line change
Expand Up @@ -6,7 +6,7 @@ FILES=warcs1/*
for f in $FILES
do
echo "Processing $f file..."
java -Dfile.encoding=UTF-8 -Xmx1024M -Djava.io.tmpdir=tika_tmp -jar warc-indexer-3.2.0-SNAPSHOT-jar-with-dependencies.jar -c config3.conf -s "http://localhost:8983/solr/netarchivebuilder" $f
java -Dfile.encoding=UTF-8 -Xmx1024M -Djava.io.tmpdir=tika_tmp -jar warc-indexer-3.3.1-jar-with-dependencies.jar -c config3.conf -s "http://localhost:8983/solr/netarchivebuilder" $f
done

echo "Flushing Solr. Documents will be visible after flush"
Expand Down
2 changes: 1 addition & 1 deletion src/bundle/indexing/batch_warcs2_folder.bat
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
cd /D "%~dp0"

FOR /R warcs2 %%G IN (*.*) do java -Dfile.encoding=UTF-8 -Xmx2048M -Djava.io.tmpdir=tika_tmp -jar warc-indexer-3.2.0-SNAPSHOT-jar-with-dependencies.jar -c config3.conf -s "http://localhost:8983/solr/netarchivebuilder" "%%G"
FOR /R warcs2 %%G IN (*.*) do java -Dfile.encoding=UTF-8 -Xmx2048M -Djava.io.tmpdir=tika_tmp -jar warc-indexer-3.3.1-jar-with-dependencies.jar -c config3.conf -s "http://localhost:8983/solr/netarchivebuilder" "%%G"

2 changes: 1 addition & 1 deletion src/bundle/indexing/batch_warcs2_folder.sh
Original file line number Diff line number Diff line change
Expand Up @@ -6,7 +6,7 @@ FILES=warcs2/*
for f in $FILES
do
echo "Processing $f file..."
java -Dfile.encoding=UTF-8 -Xmx1024M -Djava.io.tmpdir=tika_tmp -jar warc-indexer-3.2.0-SNAPSHOT-jar-with-dependencies.jar -c config3.conf -s "http://localhost:8983/solr/netarchivebuilder" $f
java -Dfile.encoding=UTF-8 -Xmx1024M -Djava.io.tmpdir=tika_tmp -jar warc-indexer-3.3.1-jar-with-dependencies.jar -c config3.conf -s "http://localhost:8983/solr/netarchivebuilder" $f
done

echo "Flushing Solr. Documents will be visible after flush"
Expand Down
2 changes: 1 addition & 1 deletion src/bundle/indexing/show_warc_config.sh
Original file line number Diff line number Diff line change
Expand Up @@ -2,5 +2,5 @@

pushd ${BASH_SOURCE%/*} > /dev/null

java -cp warc-indexer-3.2.0-SNAPSHOT-jar-with-dependencies.jar uk.bl.wa.util.ConfigPrinter
java -cp warc-indexer-3.3.1-jar-with-dependencies.jar uk.bl.wa.util.ConfigPrinter

2 changes: 1 addition & 1 deletion src/bundle/indexing/warc-indexer.sh
Original file line number Diff line number Diff line change
Expand Up @@ -30,7 +30,7 @@ THREADS_DEFAULT="2"

: ${WARCS:="$@"}

: ${INDEXER_JAR:="${WI_HOME}/warc-indexer-3.2.0-SNAPSHOT-jar-with-dependencies.jar"}
: ${INDEXER_JAR:="${WI_HOME}/warc-indexer-3.3.1-jar-with-dependencies.jar"}
: ${INDEXER_MEM:="1024M"}
: ${INDEXER_CONFIG:="${WI_HOME}/config3.conf"}
: ${INDEXER_CUSTOM:=""} # Custom arguments, e.g. "--collection OurHeritage2020" goes here
Expand Down

0 comments on commit f9ca772

Please sign in to comment.