Class: Core::ElasticConnectorActions

Inherits:
Object
  • Object
show all
Defined in:
lib/core/elastic_connector_actions.rb

Class Method Summary collapse

Class Method Details

.connector_with_concurrency_control(connector_id) ⇒ Object 



201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
 
# File 'lib/core/elastic_connector_actions.rb', line 201

def connector_with_concurrency_control(connector_id)
  seq_no = nil
  primary_term = nil

  doc = client.get(
    :index => Utility::Constants::CONNECTORS_INDEX,
    :id => connector_id,
    :ignore => 404,
    :refresh => true
  ).tap do |response|
    seq_no = response['_seq_no']
    primary_term = response['_primary_term']
  end

  { doc: doc, seq_no: seq_no, primary_term: primary_term }
end

.connectors_meta ⇒ Object 



62
63
64
65
66
67
68
69
70
71
72
 
# File 'lib/core/elastic_connector_actions.rb', line 62

def connectors_meta
  # TODO: remove the usage of with_indifferent_access. Ideally this should return a hash or nil if not found
  alias_mappings = client.indices.get_mapping(:index => Utility::Constants::CONNECTORS_INDEX, :ignore => 404).with_indifferent_access
  index = get_latest_index_in_alias(Utility::Constants::CONNECTORS_INDEX, alias_mappings.keys)
  alias_mappings.dig(index, 'mappings', '_meta') || {
    :extract_binary_content => true,
    :name => 'ent-search-generic-ingestion',
    :reduce_whitespace => true,
    :run_ml_inference => false,
  }
end

.convert_connector_filtering_to_job_filtering(connector_filtering) ⇒ Object 



242
243
244
245
246
247
248
249
250
251
252
253
254
 
# File 'lib/core/elastic_connector_actions.rb', line 242

def convert_connector_filtering_to_job_filtering(connector_filtering)
  return [] unless connector_filtering
  connector_filtering = [connector_filtering] unless connector_filtering.is_a?(Array)
  connector_filtering.each_with_object([]) do |filtering_domain, job_filtering|
    snippet = filtering_domain.dig('active', 'advanced_snippet') || {}
    job_filtering << {
      'domain' => filtering_domain['domain'],
      'rules' => filtering_domain.dig('active', 'rules'),
      'advanced_snippet' => snippet['value'] || snippet,
      'warnings' => [] # TODO: in https://github.com/elastic/enterprise-search-team/issues/3174
    }
  end
end

.create_connector(index_name, service_type) ⇒ Object 



42
43
44
45
46
47
48
49
50
 
# File 'lib/core/elastic_connector_actions.rb', line 42

def create_connector(index_name, service_type)
  body = {
    :scheduling => { :enabled => true },
    :index_name => index_name,
    :service_type => service_type
  }
  response = client.index(:index => Utility::Constants::CONNECTORS_INDEX, :body => body)
  response['_id']
end

.create_job(connector_settings:) ⇒ Object

Raises:



218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
 
# File 'lib/core/elastic_connector_actions.rb', line 218

def create_job(connector_settings:)
  body = {
    status: Connectors::SyncStatus::PENDING,
    created_at: Time.now,
    last_seen: Time.now,
    trigger_method: connector_settings.sync_now? ? Connectors::JobTriggerMethod::ON_DEMAND : Connectors::JobTriggerMethod::SCHEDULED,
    connector: {
      id: connector_settings.id,
      filtering: convert_connector_filtering_to_job_filtering(connector_settings.filtering),
      index_name: connector_settings.index_name,
      language: connector_settings[:language],
      pipeline: connector_settings[:pipeline],
      service_type: connector_settings.service_type,
      configuration: connector_settings.configuration
    }
  }

  index_response = client.index(index: Utility::Constants::JOB_INDEX, body: body, refresh: true)

  return index_response if index_response['result'] == 'created'

  raise JobNotCreatedError.new(connector_settings.id, index_response)
end

.delete_indices(indices) ⇒ Object 



107
108
109
 
# File 'lib/core/elastic_connector_actions.rb', line 107

def delete_indices(indices)
  client.indices.delete(:index => indices, :ignore_unavailable => true)
end

.delete_jobs_by_query(query) ⇒ Object 



100
101
102
103
104
105
 
# File 'lib/core/elastic_connector_actions.rb', line 100

def delete_jobs_by_query(query)
  client.delete_by_query(
    :index => Utility::Constants::JOB_INDEX,
    :body => { :query => query }
  )
end

.disable_connector_scheduling(connector_id) ⇒ Object 



120
121
122
123
 
# File 'lib/core/elastic_connector_actions.rb', line 120

def disable_connector_scheduling(connector_id)
  payload = { :enabled => false }
  update_connector_fields(connector_id, :scheduling => payload)
end

.document_count(index_name) ⇒ Object 



546
547
548
549
 
# File 'lib/core/elastic_connector_actions.rb', line 546

def document_count(index_name)
  client.indices.refresh(:index => index_name, :ignore_unavailable => true)
  client.count(:index => index_name, :ignore_unavailable => true)['count']
end

.enable_connector_scheduling(connector_id, cron_expression) ⇒ Object 



115
116
117
118
 
# File 'lib/core/elastic_connector_actions.rb', line 115

def enable_connector_scheduling(connector_id, cron_expression)
  payload = { :enabled => true, :interval => cron_expression }
  update_connector_fields(connector_id, :scheduling => payload)
end

.ensure_connectors_index_exists ⇒ Object

DO NOT USE this method Creation of connector index should be handled by Kibana, this method is only used by ftest.rb



345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
 
# File 'lib/core/elastic_connector_actions.rb', line 345

def ensure_connectors_index_exists
  mappings = {
    :dynamic => false,
    :properties => {
      :api_key_id => { :type => :keyword },
      :configuration => { :type => :object },
      :custom_schedule => { :type => :object },
      :description => { :type => :text },
      :error => { :type => :keyword },
      :features => {
        :properties => {
          :filtering_advanced_config => { :type => :boolean },
          :filtering_rules => { :type => :boolean }
        }
      },
      :filtering => {
        :properties => {
          :domain => { :type => :keyword },
          :active => {
            :properties => {
              :rules => {
                :properties => {
                  :id => { :type => :keyword },
                  :policy => { :type => :keyword },
                  :field => { :type => :keyword },
                  :rule => { :type => :keyword },
                  :value => { :type => :keyword },
                  :order => { :type => :short },
                  :created_at => { :type => :date },
                  :updated_at => { :type => :date }
                }
              },
              :advanced_snippet => {
                :properties => {
                  :value => { :type => :object },
                  :created_at => { :type => :date },
                  :updated_at => { :type => :date }
                 }
              },
              :validation => {
                :properties => {
                  :state => { :type => :keyword },
                  :errors => {
                    :properties => {
                      :ids => { :type => :keyword },
                      :messages => { :type => :text }
                    }
                  }
                }
              }
            }
          },
          :draft => {
            :properties => {
              :rules => {
                :properties => {
                  :id => { :type => :keyword },
                  :policy => { :type => :keyword },
                  :field => { :type => :keyword },
                  :rule => { :type => :keyword },
                  :value => { :type => :keyword },
                  :order => { :type => :short },
                  :created_at => { :type => :date },
                  :updated_at => { :type => :date }
                }
              },
              :advanced_snippet => {
                :properties => {
                  :value => { :type => :object },
                  :created_at => { :type => :date },
                  :updated_at => { :type => :date }
                }
              },
              :validation => {
                :properties => {
                  :state => { :type => :keyword },
                  :errors => {
                    :properties => {
                      :ids => { :type => :keyword },
                      :messages => { :type => :text }
                    }
                  }
                }
              }
            }
          }
        }
      },
      :index_name => { :type => :keyword },
      :is_native => { :type => :boolean },
      :language => { :type => :keyword },
      :last_seen => { :type => :date },
      :last_sync_error => { :type => :keyword },
      :last_sync_status => { :type => :keyword },
      :last_synced => { :type => :date },
      :last_deleted_document_count => { :type => :long },
      :last_indexed_document_count => { :type => :long },
      :name => { :type => :keyword },
      :pipeline => {
        :properties => {
          :extract_binary_content => { :type => :boolean },
          :name => { :type => :keyword },
          :reduce_whitespace => { :type => :boolean },
          :run_ml_inference => { :type => :boolean }
        }
      },
      :scheduling => {
        :properties => {
          :enabled => { :type => :boolean },
          :interval => { :type => :text }
        }
      },
      :service_type => { :type => :keyword },
      :status => { :type => :keyword },
      :sync_now => { :type => :boolean }
    }
  }
  ensure_index_exists("#{Utility::Constants::CONNECTORS_INDEX}-v1", system_index_body(:alias_name => Utility::Constants::CONNECTORS_INDEX, :mappings => mappings))
end

.ensure_content_index_exists(index_name, use_icu_locale = false, language_code = nil) ⇒ Object 



300
301
302
303
304
305
306
 
# File 'lib/core/elastic_connector_actions.rb', line 300

def ensure_content_index_exists(index_name, use_icu_locale = false, language_code = nil)
  settings = Utility::Elasticsearch::Index::TextAnalysisSettings.new(:language_code => language_code, :analysis_icu => use_icu_locale).to_h
  mappings = Utility::Elasticsearch::Index::Mappings.default_text_fields_mappings(:connectors_index => true)

  body_payload = { settings: settings, mappings: mappings }
  ensure_index_exists(index_name, body_payload)
end

.ensure_index_exists(index_name, body = {}) ⇒ Object 



308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
 
# File 'lib/core/elastic_connector_actions.rb', line 308

def ensure_index_exists(index_name, body = {})
  if client.indices.exists?(:index => index_name)
    return unless body[:mappings]
    Utility::Logger.debug("Index #{index_name} already exists. Checking mappings...")
    Utility::Logger.debug("New mappings: #{body[:mappings]}")
    response = client.indices.get_mapping(:index => index_name)
    existing = response[index_name]['mappings']
    if existing.empty?
      Utility::Logger.debug("Index #{index_name} has no mappings. Adding mappings...")
      client.indices.put_mapping(:index => index_name, :body => body[:mappings], :expand_wildcards => 'all')
      Utility::Logger.debug("Index #{index_name} mappings added.")
    else
      Utility::Logger.debug("Index #{index_name} already has mappings: #{existing}. Skipping...")
    end
  else
    client.indices.create(:index => index_name, :body => body)
    Utility::Logger.debug("Created index #{index_name}")
  end
end

.ensure_job_index_exists ⇒ Object

DO NOT USE this method Creation of job index should be handled by Kibana, this method is only used by ftest.rb



467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
 
# File 'lib/core/elastic_connector_actions.rb', line 467

def ensure_job_index_exists
  mappings = {
    :dynamic => false,
    :properties => {
      :cancelation_requested_at => { :type => :date },
      :canceled_at => { :type => :date },
      :completed_at => { :type => :date },
      :connector => {
        :properties => {
          :configuration => { :type => :object },
          :filtering => {
            :properties => {
              :domain => { :type => :keyword },
              :rules => {
                :properties => {
                  :id => { :type => :keyword },
                  :policy => { :type => :keyword },
                  :field => { :type => :keyword },
                  :rule => { :type => :keyword },
                  :value => { :type => :keyword },
                  :order => { :type => :short },
                  :created_at => { :type => :date },
                  :updated_at => { :type => :date }
                }
              },
              :advanced_snippet => {
                :properties => {
                  :value => { :type => :object },
                  :created_at => { :type => :date },
                  :updated_at => { :type => :date }
                }
              },
              :warnings => {
                :properties => {
                  :ids => { :type => :keyword },
                  :messages => { :type => :text }
                }
              }
            }
          },
          :id => { :type => :keyword },
          :index_name => { :type => :keyword },
          :language => { :type => :keyword },
          :pipeline => {
            :properties => {
              :extract_binary_content => { :type => :boolean },
              :name => { :type => :keyword },
              :reduce_whitespace => { :type => :boolean },
              :run_ml_inference => { :type => :boolean }
            }
          },
          :service_type => { :type => :keyword }
        }
      },
      :created_at => { :type => :date },
      :deleted_document_count => { :type => :integer },
      :error => { :type => :text },
      :indexed_document_count => { :type => :integer },
      :indexed_document_volume => { :type => :integer },
      :last_seen => { :type => :date },
      :metadata => { :type => :object },
      :started_at => { :type => :date },
      :status => { :type => :keyword },
      :total_document_count => { :type => :integer },
      :trigger_method => { :type => :keyword },
      :worker_hostname => { :type => :keyword }
    }
  }
  ensure_index_exists("#{Utility::Constants::JOB_INDEX}-v1", system_index_body(:alias_name => Utility::Constants::JOB_INDEX, :mappings => mappings))
end

.fetch_document_ids(index_name) ⇒ Object 



267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
 
# File 'lib/core/elastic_connector_actions.rb', line 267

def fetch_document_ids(index_name)
  page_size = 1000
  result = []
  begin
    pit_id = client.open_point_in_time(:index => index_name, :keep_alive => '1m', :expand_wildcards => 'all')['id']
    body = {
      :query => { :match_all => {} },
      :sort => [{ :id => { :order => :asc } }],
      :pit => {
        :id => pit_id,
        :keep_alive => '1m'
      },
      :size => page_size,
      :_source => false
    }
    loop do
      response = client.search(:body => body)
      hits = response.dig('hits', 'hits') || []

      ids = hits.map { |h| h['_id'] }
      result += ids
      break if hits.size < page_size

      body[:search_after] = hits.last['sort']
      body[:pit][:id] = response['pit_id']
    end
  ensure
    client.close_point_in_time(:index => index_name, :body => { :id => pit_id })
  end

  result
end

.force_sync(connector_id) ⇒ Object 



38
39
40
 
# File 'lib/core/elastic_connector_actions.rb', line 38

def force_sync(connector_id)
  update_connector_fields(connector_id, :scheduling => { :enabled => true }, :sync_now => true)
end

.get_connector(connector_id) ⇒ Object 



52
53
54
55
 
# File 'lib/core/elastic_connector_actions.rb', line 52

def get_connector(connector_id)
  # TODO: remove the usage of with_indifferent_access. Ideally this should return a hash or nil if not found
  client.get(:index => Utility::Constants::CONNECTORS_INDEX, :id => connector_id, :ignore => 404).with_indifferent_access
end

.get_job(job_id) ⇒ Object 



57
58
59
60
 
# File 'lib/core/elastic_connector_actions.rb', line 57

def get_job(job_id)
  # TODO: remove the usage of with_indifferent_access. Ideally this should return a hash or nil if not found
  client.get(:index => Utility::Constants::JOB_INDEX, :id => job_id, :ignore => 404).with_indifferent_access
end

.search_connectors(query, page_size, offset) ⇒ Object 



74
75
76
77
78
79
80
81
82
83
84
85
 
# File 'lib/core/elastic_connector_actions.rb', line 74

def search_connectors(query, page_size, offset)
  client.search(
    :index => Utility::Constants::CONNECTORS_INDEX,
    :ignore => 404,
    :body => {
      :size => page_size,
      :from => offset,
      :query => query,
      :sort => ['name']
    }
  )
end

.search_jobs(query, page_size, offset) ⇒ Object 



87
88
89
90
91
92
93
94
95
96
97
98
 
# File 'lib/core/elastic_connector_actions.rb', line 87

def search_jobs(query, page_size, offset)
  client.search(
    :index => Utility::Constants::JOB_INDEX,
    :ignore => 404,
    :body => {
        :size => page_size,
        :from => offset,
        :query => query,
        :sort => ['created_at']
    }
  )
end

.set_configurable_field(connector_id, field_name, label, value) ⇒ Object 



125
126
127
128
 
# File 'lib/core/elastic_connector_actions.rb', line 125

def set_configurable_field(connector_id, field_name, label, value)
  payload = { field_name => { :value => value, :label => label } }
  update_connector_configuration(connector_id, payload)
end

.system_index_body(alias_name: nil, mappings: nil) ⇒ Object 



328
329
330
331
332
333
334
335
336
337
338
339
340
341
 
# File 'lib/core/elastic_connector_actions.rb', line 328

def system_index_body(alias_name: nil, mappings: nil)
  body = {
    :settings => {
      :index => {
        :hidden => true,
        :number_of_replicas => 0,
        :auto_expand_replicas => '0-5'
      }
    }
  }
  body[:aliases] = { alias_name => { :is_write_index => true } } unless alias_name.nil? || alias_name.empty?
  body[:mappings] = mappings unless mappings.nil?
  body
end

.update_connector_configuration(connector_id, configuration) ⇒ Object 



111
112
113
 
# File 'lib/core/elastic_connector_actions.rb', line 111

def update_connector_configuration(connector_id, configuration)
  update_connector_fields(connector_id, :configuration => configuration)
end

.update_connector_custom_scheduling_last_synced(connector_id, schedule_key) ⇒ Object 



182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
 
# File 'lib/core/elastic_connector_actions.rb', line 182

def update_connector_custom_scheduling_last_synced(connector_id, schedule_key)
  doc = connector_with_concurrency_control(connector_id)

  body = {
    :custom_scheduling => {
      schedule_key => {
        :last_synced => Time.now
      }
    }
  }

  update_connector_fields(
    connector_id,
    body,
    doc[:seq_no],
    doc[:primary_term]
  )
end

.update_connector_fields(connector_id, doc = {}, seq_no = nil, primary_term = nil) ⇒ Object 



538
539
540
 
# File 'lib/core/elastic_connector_actions.rb', line 538

def update_connector_fields(connector_id, doc = {}, seq_no = nil, primary_term = nil)
  update_doc_fields(Utility::Constants::CONNECTORS_INDEX, connector_id, doc, seq_no, primary_term)
end

.update_connector_status(connector_id, status, error_message = nil) ⇒ Object 



256
257
258
259
260
261
262
263
264
265
 
# File 'lib/core/elastic_connector_actions.rb', line 256

def update_connector_status(connector_id, status, error_message = nil)
  if status == Connectors::ConnectorStatus::ERROR && error_message.nil?
    raise ArgumentError, 'error_message is required when status is error'
  end
  body = {
    :status => status,
    :error => status == Connectors::ConnectorStatus::ERROR ? error_message : nil
  }
  update_connector_fields(connector_id, body)
end

.update_connector_sync_now(connector_id, sync_now) ⇒ Object 



152
153
154
155
156
157
158
159
160
161
162
163
 
# File 'lib/core/elastic_connector_actions.rb', line 152

def update_connector_sync_now(connector_id, sync_now)
  doc = connector_with_concurrency_control(connector_id)

  body = { sync_now: sync_now, last_synced: Time.now }

  update_connector_fields(
    connector_id,
    body,
    doc[:seq_no],
    doc[:primary_term]
  )
end

.update_connector_sync_start(connector_id) ⇒ Object 



165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
 
# File 'lib/core/elastic_connector_actions.rb', line 165

def update_connector_sync_start(connector_id)
  doc = connector_with_concurrency_control(connector_id)

  body = {
      last_sync_status: Connectors::SyncStatus::IN_PROGRESS,
      last_sync_error: nil,
      status: Connectors::ConnectorStatus::CONNECTED
  }

  update_connector_fields(
    connector_id,
    body,
    doc[:seq_no],
    doc[:primary_term]
  )
end

.update_filtering_validation(connector_id, filter_validation_results) ⇒ Object 



130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
 
# File 'lib/core/elastic_connector_actions.rb', line 130

def update_filtering_validation(connector_id, filter_validation_results)
  return if filter_validation_results.empty?

  filtering = get_connector(connector_id).dig(:_source, :filtering)

  case filtering
  when Hash
    update_filter_validation(filtering, filter_validation_results)
  when Array
    return unless should_update_validations?(filter_validation_results, filtering)

    filtering.each do |filter|
      update_filter_validation(filter, filter_validation_results)
    end
  else
    Utility::Logger.warn("Elasticsearch returned invalid filtering format: #{filtering}. Skipping validation.")
    return
  end

  update_connector_fields(connector_id, { :filtering => filtering })
end

.update_job_fields(job_id, doc = {}, seq_no = nil, primary_term = nil) ⇒ Object 



542
543
544
 
# File 'lib/core/elastic_connector_actions.rb', line 542

def update_job_fields(job_id, doc = {}, seq_no = nil, primary_term = nil)
  update_doc_fields(Utility::Constants::JOB_INDEX, job_id, doc, seq_no, primary_term)
end