DPUL

A Spotlight application for Princeton University Library, formerly known as Pomegranate, but renamed to the offical brand: Digital PUL.

Dependencies

• Ruby
• Nodejs
• Java (to run Solr server)
• Postgres
• Redis

Initial Setup

git clone git@github.com:pulibrary/dpul.git
cd dpul
bundle install
yarn install

Remember you'll need to run bundle install and yarn install on an ongoing basis as dependencies are updated.

Setup server

Lando

Lando will automatically set up docker images for Solr and Postgres which match the versions we use in Production. The ports will not collide with any other projects you're using Solr/Postgres for, and you can easily clean up with lando destroy or turn off all services with lando poweroff.

1. Install Lando DMG from https://github.com/lando/lando/releases
2. rake servers:start

Running Tests

bundle exec rspec

Running in Development

1. Run each of the services listed below in its own terminal as necessary.

   backend: bin/rails s
   frontend: bin/vite dev
   sidekiq: bundle exec sidekiq

2. Access DPUL at http://localhost:3000/

Importing Development Data:

1. Log in once via CAS
2. Run rake dpul:site_admin
3. Click "Create a New Collection"
4. Select a small collection and hit "Save"
   • To find a small collection: go to Figgy, submit a blank search, open the facet collection and click 'more', and page to the low-count collections
5. Either wait for a solr commit or manually commit in the rails console with Blacklight.default_index.connection.commit

Auto-update from Figgy

Figgy announces events to a durable RabbitMQ fanout exchange. In order to use them, do the following:

1. Configure the events settings in config/config.yml
2. Run WORKERS=FiggyEventHandler rake sneakers:run

This will subscribe to the events and update the DPUL records when they're created, updated, or deleted.

Reindexing production or staging

To ensure production or staging indexes are up to date:

1. ssh as the deploy user
2. cd /opt/dpul/current
3. bundle exec rake dpul:reindex:collections

Note that this enqueues indexing for each collection, and actual indexing is run via background jobs. On DPUL, indexing a collection also deletes records that have been removed from the collection.

Replicate database and solr index from production to staging

• Note it will default to today's backups unless you supply an env var like DATE=2021-10-21

ssh pulsys@dpul-staging1 'sudo service nginx stop' && bundle exec cap staging replicate:to_staging && ssh pulsys@dpul-staging1 'sudo service nginx start'