# This is the default configuration file for pull indexing. # Please see plugins/generic/lucene/README for more information # about indexing modes and embedded vs. central server deployment. # # The following default configuration works unchanged when # working with an embedded solr server that indexes articles # from a single OJS installation. Please read the inline # comments for hints how to change the file in case you # index articles from several OJS installations or when you # work in central server mode. # The URL(s) where Solr should look for changed articles. If you # want to collect articles from several OJS instances then separate # the URLs by pipe symbols (|). Depending on your URL scheme you'll # have to add '/index.php' or not (path-based URLs). PULL_ENDPOINTS="$OJS_BASE_URL/index.php" # The DIH endpoint URL points to the request handler on the Solr # server that manages document upload. We assume by default that # you're using an embedded deployment. If you have a central server # deployment then you'll have to change this URL. Please change # the HTTP BASIC password to the correct value. DIH_ENDPOINT="http://admin:please change@localhost:8983/solr/ojs/dih" # The following directories are used to maintain file processing # queues. You can set them to whatever directory writable by the # user that executes the pull.sh and load.sh scripts. # The staging directory contains all files that have been downloaded # from OJS instances but have not yet been processed. PULL_STAGING_DIR="$OJS_FILES/lucene/pull/staging" # The directory for rejected files contains files that could not be # processed by the Solr import process. If you see files piling up # there then you have to manually check logs and find out what was # going wrong. After correcting the error you may simple move the # files back to the staging directory and they'll automatically be # processed. PULL_REJECT_DIR="$OJS_FILES/lucene/pull/rejected" # The archive directory contains all files that have been successfully # processed. You should compress and/or move those files on a regular # basis. You should not delete those files if you want to use them # to re-play updates after an index crash. To do so, restore the index # from a backup to the last state known to be good. Then move all files # archived after the backup timestamp to the staging queue and they'll # be automatically re-played. PULL_ARCHIVE_DIR="$OJS_FILES/lucene/pull/archive"