Pumpkin. A Hydra head to support digitization workflows

Pumpkin is a Hydra head based on Plum and CurationConcerns, with two types of works:

• ScannedResource: a book or other resource composed of one or more scanned pages
• MultiVolumeWork: a book set, sammelband or other resource composed of multiple ScannedResources

Features

• Drag-and-drop tools for reordering FileSets and editing structure
• Generating IIIF manifests for Collections and Works based on that structure
• Building PDFs of Works based on their IIIF manifests
• Performing OCR with Tesseract
• Simple state-based workflow
• Retrieving external metadata from our finding aids and catalog web services

Known Issues

• When ingesting or uploading files, the filename cannot have spaces.

Dependencies

• Redis
  • Start Redis with redis-server or if you're on certain Linuxes, you can do this via sudo service redis-server start.
• Kakadu
  • The installer doesn't work on MacOSX 10.11 (El Capitan), but the files kdu_compress and libkdu_v77R.dylib can be extracted from the download packages and used by manually installing them in /usr/local/bin and /usr/local/lib respectively.
• Tesseract
  • Version 3.04 is required. You can install it on Mac OSX with brew install tesseract --with-all-languages For Ubuntu you'll have to compile it.
• RabbitMQ (Optional)
  • Start with rabbitmq-server
  • Used for publishing create/update/delete events for systems such as Pomegranate

Initial Setup

You may need to prefix rake commands with bundle exec, particularly if you have a newer version of the rake gem installed.

• Install dependencies: bundle install
• Setup the database: rake db:migrate
• Setup ActiveFedora::Noid minter: rails g active_fedora:noid:seed

Running the Tests

Setup dependencies and run the test suite:

$ bundle install
$ rake db:migrate
$ rake ci

You may need to create the tmp directory, which can be done automatically by starting the rails server (rails s) and then stopping it.

You may also want to run the Fedora and Solr servers in one window with:

$ rake hydra:test_server

And run the test suite in another window:

$ rake spec

Running the Services

Pumpkin consists of several cooperating services: Fedora, Solr, a Blacklight (i.e. Rails) application, Redis, and an AMQP 0.9 broker such as RabbitMQ.

Redis and the broker (if used) are assumed to be running separately, as noted under Dependencies above.

For development and testing, Pumpkin includes wrapper scripts for Fedora and Solr. These will download the services as needed and start them for you. Rake tasks to configure and run the wrappers are included, as detailed below. You can stop these supporting services by typing the interrupt character (i.e. control-C) in the console where you are running the wrappers task.

To start Fedora and Solr services to support development:

rake server:development

They can be configured by editing .fcrepo_wrapper and .solr_wrapper.

To start Fedora and Solr to support testing:

rake hydra:test_server

They can be configured by editing config/fcrepo_wrapper_test.yml and config/solr_wrapper_test.yml.

In either case, you can then start Blacklight in the normal fashion for Rails applications:

rails server

or

rails console

The Pumpkin user interface should then be available at http://localhost:3000/

To start Fedora, Solr and Blacklight on the console:

rake hydra:server

Production deployment will vary depending on your local procedures and requirements.

Please note that config/fedora.yml and config/solr.yml are for configuring the Blacklight application's connections to Fedora and Solr, which is separate from configuring instances of Fedora and Solr themselves.

Adding an Admin user

1. Run the development servers with rake server:development
2. Run Plum with rails s
3. Go to http://localhost:3000/users/auth/cas and login with CAS
4. $ rake add_admin_role

Configuring Loris for Development

1. Install Docker Toolbox https://www.docker.com/toolbox

• Only necessary for mac or windows machines. For unix boxes install via wget -qO- https://get.docker.com/ | sh

1. Start a docker VM: docker-machine start default
2. Setup your docker environment: eval "$(docker-machine env default)"
3. Retrieve the loris image: docker pull lorisimageserver/loris
4. Start the container:

docker run --name loris -v /path/to/plum/tmp/derivatives:/usr/local/share/images -d -p 5004:5004 lorisimageserver/loris

1. Find the docker IP address with docker-machine ls
2. Export config variable for IIIF url: export PLUM_IIIF_URL="http://<docker-ip>:5004"
3. Images should be available at http://<docker-ip>:5004/ based on the FileSet id. e.g., if your docker IP address is 192.168.99.100, the full view of FileSet 70795765b would be at http://192.168.99.100:5004/70%2F79%2F%2F57%2F65%2Fb-intermediate_file.jp2/full/full/0/default.jpg