Iris

This project is a Hyrax 1.0.5 application that uses the Samvera Labs geo_works gem to ingest works with spatial geographical data.

Functional Spec

A functional spec of DCE's first phase of GIS work for UCSB

Developer Setup

1. Change to your working directory for new development projects cd .
2. Clone this repo git clone git@github.com:curationexperts/iris.git For UCSB Library repo version git clone https://github.com/ucsblibrary/iris.git
3. Change to the application directory cd iris
4. bundle install under project's current ruby (2.4.4)
5. Start redis redis-server & note: use & to start in the background, or run redis in a new terminal session

Integration with a geoblacklight-based app

If you want your fedora objects to be indexed in another solr instance that is used by a geoblacklight app, set the environment variable to point to that solr instance:

export GEOBLACKLIGHT_SOLR_URL="http://127.0.0.1:8987/solr/development"

Indexing into the geoblacklight solr happens in a background job, so make sure your environment is configured to run background jobs.

Integration with GeoServer

See notes in the README_geoserver.md

Install Geo_works dependencies

1. GDAL

• Mac OSX: brew install gdal
• Ubuntu: use sudo apt-get install gdal-bin

2. Simple Tiles

• Mac OS X: brew install simple-tiles

• Linux:

  Install dependencies:

  ruby
  libgdal-dev
  libcairo2-dev
  libpango1.0-dev
  

  Compile:

  ruby
  $ git clone git@github.com:propublica/simple-tiles.git
  $ cd simple-tiles
  $ ./configure
  $ make && make install
  

Create db

bundle exec rake db:setup bundle exec rake db:migrate

Create default admin set and load workflows

1. Start solr and fedora (in new tabs) bundle exec solr_wrapper and bundle exec fcrepo_wrapper.

2. Ensure AdminSet dependencies are in place: bundle exec rake hyrax:workflow:load and bundle exec rake hyrax:default_admin_set:create

3. Start server bundle exec rails s

Run the app in development mode

• Start redis server
• Start tomcat (if you want to run GeoServer)
• Run fcrepo_wrapper from my iris workspace
• Run solr_wrapper from my iris workspace
• Run solr_wrapper from my aster workspace (if you want to integrate with the geoblacklight app)
• Start sidekiq (if you want to run background jobs)
• Start rails s

Import works using the command-line importer

Set the following environment variables:

export RAILS_HOST='localhost:3000'

To import work records from a directory that contains zip files:

rake iris:import:from_zips[/path/to/your/zips]

Notes and Assumptions about Command-Line Importer:

• The importer expects all the binaries to be found in zip files, for both Raster and Vector works (even for files that are not normally multi-part files, such as GeoTIFF).

• You can use the importer to create new records, but currently, you cannot update existing records by re-importing them.

• The importer queues a set of background jobs for each zip file it processes, and some jobs queue other jobs. You'll need to wait for the background jobs to finish running before you will see the new data appear in your GeoBlacklight app and your GeoServer.

• For the first release, the importer has only been tested with GeoTIFF and Shapefile. Support for other types of Vector or Raster works will be a future round of work.

• For the first release, the importer has only been tested with records that should be publicly available. Support for login-restricted works will be a future round of work.

• Although the geo_works gem allows 3 work types, VectorWork, RasterWork, and ImageWork, the importer currently cannot create ImageWork records.

• Although the geo_works gem allows 2 types of Grid file (ASCII and binary), the importer currently only imports binary Grids.