Module: ElasticSearch::AggregatableRelation

Included in:
AggregationRelation, Relation
Defined in:
lib/elastic_search/aggregatable_relation.rb

Overview

The ElasticSearch::AggregatableRelation mixin provides handy methods for using the ElasticSearch aggregation framework, which can be chained with each other, all other relation methods and even nested.

Examples:

ProductIndex.where(available: true).aggregate(:tags, size: 50)
OrderIndex.aggregate(revenue: { sum: { field: "price" }})

Class Method Summary collapse

Instance Method Summary collapse

Class Method Details

.included(base) ⇒ Object



12
13
14
15
16
# File 'lib/elastic_search/aggregatable_relation.rb', line 12

def self.included(base)
  base.class_eval do
    attr_accessor :aggregation_values
  end
end

Instance Method Details

#aggregate(field_or_hash, options = {}, &block) ⇒ Object

Adds an arbitrary aggregation to the request which can be chained as well as nested. Check out the examples and ElasticSearch docs for further details.

Examples:

Basic usage with optons

query = CommentIndex.where(public: true).aggregate(:user_id, size: 100)

query.aggregations(:user_id)
# => { 4 => #<ElasticSearch::Result ...>, 7 => #<ElasticSearch::Result ...>, ... }

Simple range aggregation

ranges = [{ to: 50 }, { from: 50, to: 100 }, { from: 100 }]

ProductIndex.aggregate(price_range: { range: { field: "price", ranges: ranges }})

Basic nested aggregation

# When nesting aggregations, the return value of the aggregate block is
# used.

OrderIndex.aggregate(:user_id, order: { revenue: "desc" }) do |aggregation|
  aggregation.aggregate(revenue: { sum: { field: "price" }})
end

Nested histogram aggregation


OrderIndex.aggregate(histogram: { date_histogram: { field: "price", interval: "month" }}) do |aggregation|
  aggregation.aggregate(:user_id)
end

Nested aggregation with filters

OrderIndex.aggregate(average_price: {}) do |aggregation|
  aggregation = aggregation.match_all
  aggregation = aggregation.where(user_id: current_user.id) if current_user

  aggregation.aggregate(average_price: { avg: { field: "price" }})
end


55
56
57
58
59
60
61
62
63
64
65
66
67
# File 'lib/elastic_search/aggregatable_relation.rb', line 55

def aggregate(field_or_hash, options = {}, &block)
  fresh.tap do |relation|
    hash = field_or_hash.is_a?(Hash) ? field_or_hash : { field_or_hash => { terms: { field: field_or_hash }.merge(options) } }

    if block
      aggregation_relation = block.call(ElasticSearch::AggregationRelation.new)

      field_or_hash.is_a?(Hash) ? hash[field_or_hash.keys.first].merge!(aggregation_relation.to_hash) : hash[field_or_hash].merge!(aggregation_relation.to_hash)
    end

    relation.aggregation_values = (aggregation_values || {}).merge(hash)
  end
end