# frozen_string_literal: true require 'json' # Custom aggregations provides a way to aggregate data in a non-standard way to enhance statistics. # These aggregations will run in ElasticSearch so they must belongs to its syntax. class CustomAggregationService ## # Run any additional custom aggregations related to the given statistic type, if any ## def call(statistic_index, statistic_type, start_date, end_date, custom_query, results) if statistic_type && start_date && end_date stat_index = StatisticIndex.find_by(es_type_key: statistic_index) stat_type = StatisticType.find_by(statistic_index_id: stat_index.id, key: statistic_type) client = Elasticsearch::Model.client stat_type.statistic_custom_aggregations.each do |custom| query = sprintf(custom.query, aggs_name: custom.field, start_date: start_date, end_date: end_date) if custom_query && !custom_query.empty? # Here, a custom query was provided with the original request (eg: filter by subtype) # so we try to apply this custom filter to the current custom aggregation. # # The requested model mapping (ie. found in StatisticCustomAggregation.es_index > es_type) must have defined # these fields in the indexed json, otherwise the returned value will probably not be what is expected. # # As an implementation exemple, you can take a look at Availability (indexed as fablab/availabilities) # and witch will run custom filters on the fields 'date' and 'subType'. Other custom filters will return 0 # as they are not relevant with this kind of custom aggregation. query = JSON.parse(query) custom_query = JSON.parse(custom_query) exclude = custom_query.delete('exclude') if exclude query['query']['bool']['must_not'] = [{ term: custom_query['match'] }] else query['query']['bool']['must'].push(custom_query) end query = query.to_json end c_res = client.search(index: custom.es_index, type: custom.es_type, body: query) results['aggregations'][custom.field] = c_res['aggregations'][custom.field] end end results end end