image-service

This Grails application provides the webservices and backend for the storage of all images in the Atlas. It includes:

• Support for large images, generation of thumbnails and tile views
• Extensible key/value pair storage for image metadata
• Support for subimaging and maintaining the relationships between parent and child images
• Exif extraction
• Tile view for large images compatible with GIS Javascript clients such as LeafletJS, OpenLayers and Google Maps
• Web services for image upload
• Generate of derivative images for thumbnail presentation
• Tagging support via webservices
• Administrator console for image management
• Swagger API definition
• Integration with google analytics to monitor image usage by data resource
• Support for image storage in S3, Swift
• Support for batch uploads with AVRO

There are other related repositories to this one:

• images-client-plugin - a grails plugin to provide a Javascript based viewer to be used in other applications requiring a image viewer. This viewer is based on LeafletJS.
• image-tiling-agent - a utility to run tiling jobs for the image-service. This is intended to used on multiple machine as tiling is CPU intensive and best parallelised.
• image-loader - utility for bulk loading images into the image-service.

Upgrading from 1.0

Please see the Upgrading from 1.0 to 1.1 wiki page before upgrading an image-service 1.0 or earlier installation to the latest version.

Architecture

• Grails 6.1.0 web application ran as standalone executable jar
• Open JDK 11
• Postgres database (11 or above)
• Elastic search 7
• Debian package install

Installation

There are ansible scripts for this applications (and other ALA tools) in the ala-install project. The ansible playbook for the image-service is here

You can also run this application locally by following the instructions on its wiki page

Running it locally

Postgres

There is a docker-compose YML file that can be used to run postgres locally for local development purposes. To use run:

docker-compose -f postgres.yml up -d

And to shutdown

docker-compose -f postgres.yml kill

Elastic search

There is a docker-compose YML file that can be used to run elastic search locally for local development purposes. To use run:

docker-compose -f elastic.yml up -d

And to shutdown

docker-compose -f elastic.yml kill

Disable caching (HTTP + in-memory)

You can disable all caching for image responses and bypass the in‑memory Caffeine caches used for thumbnails and tiles by setting the following configuration property:

images:
  disableCache: true

Effects when images.disableCache = true:

• Controllers that serve image bytes (standard image endpoints and IIIF) will not return 304 Not Modified responses and will replace any cache headers with the following no‑cache headers on every response:
  • Cache-Control: no-store, no-cache, must-revalidate
  • Pragma: no-cache
  • Expires: 0
• ImageStoreService will bypass the Caffeine caches for thumbnail and tile lookups; results will not be cached in memory.

By default, images.disableCache is false.