PRN-maps-api

API for mapping the Planetary Response Network (tPRN) data results via https://github.com/zooniverse/prn-maps

Overview

This API is a proxy around s3 files that have been created for each tPRN activation event.

Each Event comprises the following steps:

1. A JSON manifest describing each tPRN event will be created and stored in s3
   • This manifest is served by this API to configure the mapping UI
   • See https://github.com/zooniverse/tprn-data-processing/tree/master/event_manifest for details on setting up a manifest
2. Before and after imagery of a PRN even is pushed to a Zooniverse project
3. Volunteers classify the uploaded data
4. Raw classification data is collated and archived to a known s3 path
5. IBCC code is run over the collated data
6. The resulting layer data is uploaded via the API to be served to the mapping UI

This API will fetch the event information stored in known s3 locations (as setup in the event manifest) and served as JSON data content types. This json data will be used by the mapping UI interface https://github.com/zooniverse/prn-maps for visualizing the results of tPRN event.

Routes

Public end points

GET /events

• List all known tPRN event manifests available in s3, including the manifest name and URL for retrieval via HTTPS

GET /events/${event_name}

• Show the event metadata for a known event name

GET /layers/${event_name}

• Show the data layers for a known event name

GET /layers/${event_name}/${version}/${layer_name}

• Show the data layer for a known event that matches the version and layer name paths

Protected end points (basic auth headers required)

POST /approved/layers/${event_name}/revert/${version}

• Revert the approved data layers for a known version and event name.
• You must supply the event_name and version values in the route.

GET /pending/layers/${event_name}

• Show the pending data layers for a known event name

POST /pending/layers/${event_name}/approve/${version}

• Approve and publish the pending data layers for a known version and event name.
• You must supply the event_name and version values in the route.

POST /upload/layers/${event_name}

• Upload a list of layer files and one metadata files to describe the list of layers. Use multipart/form-data encoding to upload the files
  • At least one layer file is expected via layers[] param
  • Exactly one metadata JSON file is expected on metadata param
• The metadata JSON file must conform to a valid schema

  {
  	"layers": [{
  		"file_name": "layer_1.csv",
  		"created_at": "2018-07-07T15:55:00.511Z"
  	},{
  		"file_name": "layer_2.csv",
  		"created_at": "2018-07-07T15:55:00.511Z"
  	}]
  }
  

Get started

Using docker and docker-compose docker-compose up

or get a shell via docker-compose run --rm prn_maps_api bash

Using your own ruby install bundle install bundle exec puma -C docker/puma.rb to run the API server in development mode

Running tests

Using docker docker-compose run --rm prn_maps_api ruby -Ilib -e 'ARGV.each { |f| require f }' ./spec/*_spec.rb

Manually via your own ruby install ruby -Ilib -e 'ARGV.each { |f| require f }' ./spec/*_spec.rb

Or use autotest for constant feedback on file system changes autotest