Relationships

This is the official documentation of the agent_ruby Ruby agent.

When relationships are defined during the customization step, Forest Admin Collections act as if the 2 Collections were natively linked at the data source level.

You may have noticed that relationships within a data source are configured out of the box, so you won't need to define those.

However, you may want to create additional intra and cross data source relationships to:

• help users navigate within your admin panel,

• create charts that use data from multiple data sources,

• let users filter, use scopes, or segment with conditions that cross data source boundaries.

Minimal example

@create_agent.customize_collection('town') do |collection|
  collection
    # Towns belong to 1 country
    .add_many_to_one_relation('country', 'Country', { foreign_key: 'country_id' })

   # Towns have 1 mayor
    .add_one_to_one_relation('myMayor', 'Mayor', { origin_key: 'town_id' })

    # Towns have multiple inhabitants
    .add_one_to_many_relation('inhabitants', 'Person', { origin_key: 'town_id' })

    # Towns electricity is supplied by power-plants which are shared with other towns
    .add_many_to_many_relation('energyProviders', 'PowerPlant', 'UtilityContract', { origin_key: 'town_id', foreign_key: 'power_plant_id' })

    # Towns have a list of honorary citizen that is retrievable through a public API
    .add_external_relation('honoraryCitizen', {
      # customizer.add_external_relation(
      schema: { 'firstName' => 'String', 'lastName' => 'String' },
      list_records: proc { |record|
        response = HTTParty.get("https://api.mytown.com/cities/#{record['id']}/honorary-citizen")
        response.body
      }
    })
end