Catalyst Traject Project

This project provides indexing for catalyst

• Create the solr index from horizon
• Create a temporary mysql table for shelfbrowse

Initially this project was part of the blacklight_rails project

Prerequisites:

java8 jruby-9.1

Getting Started

To set it up locally, install the gems with bundler

bundle install --path /vendor/bundle

Create a .env file by filling in the variable values in env-example. For the values, please check lastpass or the servers

Run as a cron

RAILS_ENV=production bundle exec rake horizon:mass_index

Manually run on Master Server

CD /opt/catalyst-traject
RAILS_ENV=production bundle exec rake horizon:mass_index

These are configuration and setup files to be used with the traject tool for indexing MARC to solr.

We also use the traject_horizon plugin to stream directly from the Horizon database, through traject, then to Solr.

We use a Gemfile to specify the gems used by our traject process (probably just the traject_horizon plugin), so use the -G arg to traject to tell it to use that Gemfile.

traject needs to run under jruby (at least for the configuration we're using), then for instance:

traject -G -c ./conf/horizon_source.rb -c ./conf/solr_connect.rb -c ./conf/horizon_index.rb

Or use rake tasks that we might write to control this process.