Class: BigqueryMigration::BigqueryWrapper

Inherits:
Object
  • Object
show all
Defined in:
lib/bigquery_migration/bigquery_wrapper.rb

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(config, opts = {}) ⇒ BigqueryWrapper

Returns a new instance of BigqueryWrapper.



21
22
23
24
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 21

def initialize(config, opts = {})
  @config = HashUtil.deep_symbolize_keys(config)
  @opts = HashUtil.deep_symbolize_keys(opts)
end

Instance Attribute Details

#configObject (readonly)

Returns the value of attribute config.



15
16
17
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 15

def config
  @config
end

Instance Method Details

#application_default_credentials_fileObject



774
775
776
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 774

def application_default_credentials_file
  @application_default_credentials_file ||= File.expand_path("~/.config/gcloud/application_default_credentials.json")
end

#auth_methodObject

compute_engine, authorized_user, service_account



757
758
759
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 757

def auth_method
  @auth_method ||= ENV['AUTH_METHOD'] || config.fetch(:auth_method, nil) || credentials[:type] || 'compute_engine'
end

#clientObject



26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 26

def client
  return @cached_client if @cached_client && @cached_client_expiration > Time.now

  client = Google::Apis::BigqueryV2::BigqueryService.new
  client.request_options.retries = retries
  client.client_options.open_timeout_sec = open_timeout_sec
  if client.request_options.respond_to?(:timeout_sec)
    client.request_options.timeout_sec = timeout_sec
  else # google-api-ruby-client >= v0.11.0
    if timeout_sec
      logger.warn { "timeout_sec is deprecated in google-api-ruby-client >= v0.11.0. Use read_timeout_sec instead" }
    end
    client.client_options.send_timeout_sec = send_timeout_sec
    client.client_options.read_timeout_sec = read_timeout_sec
  end
  logger.debug { "client_options: #{client.client_options.to_h}" }
  logger.debug { "request_options: #{client.request_options.to_h}" }

  scope = "https://www.googleapis.com/auth/bigquery"

  case auth_method
  when 'authorized_user'
    auth = Signet::OAuth2::Client.new(
      token_credential_uri: "https://accounts.google.com/o/oauth2/token",
      audience: "https://accounts.google.com/o/oauth2/token",
      scope: scope,
      client_id:     credentials[:client_id],
      client_secret: credentials[:client_secret],
      refresh_token: credentials[:refresh_token]
    )
    auth.refresh!
  when 'compute_engine'
    auth = Google::Auth::GCECredentials.new
  when 'service_account'
    key = StringIO.new(credentials.to_json)
    auth = Google::Auth::ServiceAccountCredentials.make_creds(json_key_io: key, scope: scope)
  when 'application_default'
    auth = Google::Auth.get_application_default([scope])
  else
    raise ConfigError, "Unknown auth method: #{auth_method}"
  end

  client.authorization = auth

  @cached_client_expiration = Time.now + 1800
  @cached_client = client
end

#clusteringObject



846
847
848
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 846

def clustering
  config[:clustering]
end

#config_defaultObject



786
787
788
789
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 786

def config_default
  # {core:{account:'xxx',project:'xxx'},compute:{zone:'xxx}}
  @config_default ||= File.readable?(config_default_file) ? HashUtil.deep_symbolize_keys(IniFile.load(config_default_file).to_h) : {}
end

#config_default_fileObject



782
783
784
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 782

def config_default_file
  File.expand_path('~/.config/gcloud/configurations/config_default')
end

#copy_table(destination_table:, destination_dataset: nil, source_table: nil, source_dataset: nil, write_disposition: nil) ⇒ Object



462
463
464
465
466
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
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 462

def copy_table(destination_table:, destination_dataset: nil, source_table: nil, source_dataset: nil, write_disposition: nil)
  source_table ||= self.table
  source_dataset ||= self.dataset
  destination_dataset ||= source_dataset
  write_disposition ||= 'WRITE_TRUNCATE'

  body = {
    job_reference: {
      project_id: self.project,
      job_id: "job_#{SecureRandom.uuid}",
    },
    configuration: {
      copy: {
        create_deposition: 'CREATE_IF_NEEDED',
        write_disposition: write_disposition,
        source_table: {
          project_id: project,
          dataset_id: source_dataset,
          table_id: source_table,
        },
        destination_table: {
          project_id: project,
          dataset_id: destination_dataset,
          table_id: destination_table,
        },
      }
    }
  }
  body[:job_reference][:location] = location if location
  opts = {}

  logger.info  { "#{head}insert_job(#{project}, #{body}, #{opts})" }
  unless dry_run?
    response = client.insert_job(project, body, opts)
    get_response = wait_load('copy', response)
  end

  {
    responses: {
      insert_job: response,
      last_get_job: get_response,
    }
  }
end

#credentialsObject



761
762
763
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 761

def credentials
  json_key || HashUtil.deep_symbolize_keys(JSON.parse(config.fetch(:credentials, nil) || File.read(credentials_file)))
end

#credentials_fileObject



765
766
767
768
769
770
771
772
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 765

def credentials_file
  @credentials_file ||= File.expand_path(
    # ref. https://developers.google.com/identity/protocols/application-default-credentials
    ENV['GOOGLE_APPLICATION_CREDENTIALS'] ||
    config.fetch(:credentials_file, nil) ||
    (File.exist?(global_application_default_credentials_file) ? global_application_default_credentials_file : application_default_credentials_file)
  )
end

#datasetObject



834
835
836
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 834

def dataset
  @dataset ||= config[:dataset] || raise(ConfigError, '`dataset` is required.')
end

#delete_table(dataset: nil, table: nil) ⇒ Object Also known as: drop_table



225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 225

def delete_table(dataset: nil, table: nil)
  dataset ||= self.dataset
  table ||= self.table

  begin
    logger.info { "#{head}Delete (drop) table... #{project}:#{dataset}.#{table}" }
    unless dry_run?
      client.delete_table(project, dataset, table) # no response
      success = true
    end
  rescue Google::Apis::ServerError, Google::Apis::ClientError, Google::Apis::AuthorizationError => e
    if e.status_code == 404 && /Not found:/ =~ e.message
      # ignore 'Not Found' error
      return {}
    end

    response = {status_code: e.status_code, message: e.message, error_class: e.class}
    raise Error, "Failed to delete_table(#{project}, #{dataset}, #{table}), response:#{response}"
  end

  { success: success }
end

#drop_column(table: nil, columns: nil, drop_columns: nil, backup_dataset: nil, backup_table: nil) ⇒ Object



594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 594

def drop_column(table: nil, columns: nil, drop_columns: nil, backup_dataset: nil, backup_table: nil)
  table ||= self.table
  backup_dataset ||= self.dataset
  if columns.nil? and drop_columns.nil?
    raise ArgumentError, '`drop_columns` or `columns` is required'
  end

  result = { responses: {} }

  before_columns = existing_columns

  if columns # if already given
    schema = Schema.new(columns)
  else
    schema = Schema.new(existing_columns)
    schema.reject_columns!(drop_columns)
  end
  if schema.empty? && !dry_run?
    raise Error, 'No column is remained'
  end

  schema.validate_permitted_operations!(before_columns)

  unless backup_dataset == self.dataset
    create_dataset(dataset: backup_dataset)
  end

  if backup_table
    _result = copy_table(source_table: table, destination_table: backup_table, destination_dataset: backup_dataset)
    result[:responses].merge!(_result[:responses])
  end

  unless (add_columns = schema.diff_columns_by_name(before_columns)).empty?
    _result = patch_table(add_columns: add_columns)
    result[:responses].merge!(_result[:responses])
  end

  query_fields = schema.build_query_fields(before_columns)
  query = "SELECT #{query_fields.join(',')} FROM [#{dataset}.#{table}]"
  _result = insert_select(query: query, destination_table: table)
  result[:responses].merge!(_result[:responses])

  after_columns = existing_columns

  result.merge!({before_columns: before_columns, after_columns: after_columns})
end

#dry_run?Boolean

Returns:

  • (Boolean)


858
859
860
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 858

def dry_run?
  @opts[:dry_run]
end

#existing_columnsObject



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

def existing_columns
  begin
    result = get_table
    response = result[:responses][:get_table]
    return [] unless response
    return [] unless response.schema
    return [] unless response.schema.fields
    response.schema.fields.map {|column| column.to_h }
  rescue NotFoundError
    return []
  end
end

#get_dataset(dataset: nil) ⇒ Object



87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 87

def get_dataset(dataset: nil)
  dataset ||= self.dataset
  begin
    logger.info { "Get dataset... #{project}:#{dataset}" }
    response = client.get_dataset(project, dataset)
  rescue Google::Apis::ServerError, Google::Apis::ClientError, Google::Apis::AuthorizationError => e
    if e.status_code == 404
      raise NotFoundError, "Dataset #{project}:#{dataset} is not found"
    end

    response = {status_code: e.status_code, message: e.message, error_class: e.class}
    raise Error, "Failed to get_dataset(#{project}, #{dataset}), response:#{response}"
  end

  { responses: { get_dataset: response } }
end

#get_table(dataset: nil, table: nil) ⇒ Object



140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 140

def get_table(dataset: nil, table: nil)
  dataset ||= self.dataset
  table ||= self.table
  begin
    logger.debug { "Get table... #{project}:#{dataset}.#{table}" }
    response = client.get_table(project, dataset, table)
  rescue Google::Apis::ServerError, Google::Apis::ClientError, Google::Apis::AuthorizationError => e
    if e.status_code == 404 # not found
      raise NotFoundError, "Table #{project}:#{dataset}.#{table} is not found"
    end

    response = {status_code: e.status_code, message: e.message, error_class: e.class}
    raise Error, "Failed to get_table(#{project}, #{dataset}, #{table}), response:#{response}"
  end

  result = {}
  if response
    result = {
      table_id: response.id,
      creation_time: response.creation_time.to_i, # millisec
      last_modified_time: response.last_modified_time.to_i, # millisec
      location: response.location,
      num_bytes: response.num_bytes.to_i,
      num_rows: response.num_rows.to_i,
    }
  end

  result.merge!({ responses: { get_table: response } })
end

#global_application_default_credentials_fileObject



778
779
780
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 778

def global_application_default_credentials_file
  @global_application_default_credentials_file ||= '/etc/google/auth/application_default_credentials.json'
end

#headObject



862
863
864
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 862

def head
  dry_run? ? '(DRY-RUN) ' : '(EXECUTE) '
end

#insert_all_table_data(dataset: nil, table: nil, rows:) ⇒ Object

rows:

- id: 1
  type: one
  record:
    child1: 'child1'
    child2: 'child2'
- id: 2
  type: two
  record:
    child1: 'child3'
    child2: 'child4'


318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 318

def insert_all_table_data(dataset: nil, table: nil, rows: )
  dataset ||= self.dataset
  table ||= self.table

  begin
    logger.info { "#{head}insertAll tableData... #{project}:#{dataset}.#{table}" }
    body = {
      rows: rows.map {|row| { json: row } },
    }
    opts = {}
    unless dry_run?
      response = client.insert_all_table_data(project, dataset, table, body, opts)
    end
  rescue Google::Apis::ServerError, Google::Apis::ClientError, Google::Apis::AuthorizationError => e
    if e.status_code == 404 # not found
      raise NotFoundError, "Table #{project}:#{dataset}.#{table} is not found"
    end

    response = {status_code: e.status_code, message: e.message, error_class: e.class}
    Medjed::Bulk.logger.error {
      "insert_all_table_data(#{project}, #{dataset}, #{table}, #{opts}), response:#{response}"
    }
    raise Error, "failed to insert_all table_data #{project}:#{dataset}.#{table}, response:#{response}"
  end

  { responses: { insert_all_table_data: response } }
end

#insert_dataset(dataset: nil, reference: nil) ⇒ Object Also known as: create_dataset



104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 104

def insert_dataset(dataset: nil, reference: nil)
  dataset ||= self.dataset
  begin
    logger.info { "#{head}Insert (create) dataset... #{project}:#{dataset}" }
    hint = {}
    if reference
      response = get_dataset(reference)
      hint = { access: response.access }
    end
    body = {
      dataset_reference: {
        project_id: project,
        dataset_id: dataset,
      },
    }.merge(hint)
    body[:location] = location if location
    opts = {}

    logger.debug { "#{head}insert_dataset(#{project}, #{body}, #{opts})" }
    unless dry_run?
      response = client.insert_dataset(project, body, opts)
    end
  rescue Google::Apis::ServerError, Google::Apis::ClientError, Google::Apis::AuthorizationError => e
    if e.status_code == 409 && /Already Exists:/ =~ e.message
      # ignore 'Already Exists' error
      return {}
    end

    response = {status_code: e.status_code, message: e.message, error_class: e.class}
    raise Error, "Failed to insert_dataset(#{project}, #{body}, #{opts}), response:#{response}"
  end

  { responses: { insert_dataset: response } }
end

#insert_partitioned_table(dataset: nil, table: nil, columns:, options: {}) ⇒ Object Also known as: create_partitioned_table



219
220
221
222
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 219

def insert_partitioned_table(dataset: nil, table: nil, columns:, options: {})
  options['time_partitioning'] = {'type'=>'DAY'}
  insert_table(dataset: dataset, table: table, columns: columns, options: options)
end

#insert_select(query:, destination_table: nil, destination_dataset: nil, write_disposition: nil) ⇒ Object



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
537
538
539
540
541
542
543
544
545
546
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 507

def insert_select(query:, destination_table: nil, destination_dataset: nil, write_disposition: nil)
  destination_table   ||= self.table
  destination_dataset ||= self.dataset
  write_disposition ||= 'WRITE_TRUNCATE'

  body  = {
    job_reference: {
      project_id: self.project,
      job_id: "job_#{SecureRandom.uuid}",
    },
    configuration: {
      query: {
        allow_large_results: true,
        flatten_results: false,
        write_disposition: write_disposition,
        query: query,
        destination_table: {
          project_id: self.project,
          dataset_id: destination_dataset,
          table_id: destination_table,
        },
      }
    }
  }
  body[:job_reference][:location] = location if location
  opts = {}

  logger.info { "#{head}insert_job(#{project}, #{body}, #{opts})" }
  unless dry_run?
    response = client.insert_job(project, body, opts)
    get_response = wait_load('query', response)
  end

  {
    responses: {
      insert_job: response,
      last_get_job: get_response,
    }
  }
end

#insert_table(dataset: nil, table: nil, columns:, options: {}) ⇒ Object Also known as: create_table

Raises:



170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 170

def insert_table(dataset: nil, table: nil, columns:, options: {})
  dataset ||= self.dataset
  table ||= self.table
  raise Error, "columns is empty" if columns.empty?
  schema = Schema.new(columns)

  begin
    logger.info { "#{head}Insert (create) table... #{project}:#{dataset}.#{table}" }
    body = {
      table_reference: {
        table_id: table,
      },
      schema: {
        fields: schema,
      }
    }

    if options['time_partitioning']
      body[:time_partitioning] = {
        type: options['time_partitioning']['type'],
        expiration_ms: options['time_partitioning']['expiration_ms'],
      }
    end

    if clustering && clustering[:fields]
      body[:clustering] = {
        fields: clustering[:fields]
      }
    end

    opts = {}
    logger.debug { "#{head}insert_table(#{project}, #{dataset}, #{body}, #{opts})" }
    unless dry_run?
      response = client.insert_table(project, dataset, body, opts)
    end
  rescue Google::Apis::ServerError, Google::Apis::ClientError, Google::Apis::AuthorizationError => e
    if e.status_code == 409 && /Already Exists:/ =~ e.message
      # ignore 'Already Exists' error
      return {}
    end

    response = {status_code: e.status_code, message: e.message, error_class: e.class}
    raise Error, "Failed to insert_table(#{project}, #{dataset}, #{body}, #{opts}), response:#{response}"
  end

  { responses: { insert_table: response } }
end

#job_status_max_polling_timeObject



854
855
856
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 854

def job_status_max_polling_time
  @job_status_max_polling_time ||= config[:job_status_polling_time] || 3600
end

#job_status_polling_intervalObject



850
851
852
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 850

def job_status_polling_interval
  @job_status_polling_interval ||= config[:job_status_polling_interval] || 5
end

#json_keyObject

For old version compatibility Use credentials_file or credentials instead



731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 731

def json_key
  if json_keyfile = config[:json_keyfile]
    begin
      case json_keyfile
      when String
        return HashUtil.deep_symbolize_keys(JSON.parse(File.read(json_keyfile)))
      when Hash
        case json_keyfile[:content]
        when String
          return HashUtil.deep_symbolize_keys(JSON.parse(json_keyfile[:content]))
        when Hash
          return json_keyfile[:content]
        else
          raise ConfigError.new "Unsupported json_keyfile type"
        end
      else
        raise ConfigError.new "Unsupported json_keyfile type"
      end
    rescue => e
      raise ConfigError.new "json_keyfile is not a JSON file"
    end
  end
  nil
end

#list_table_data(dataset: nil, table: nil, max_results: 100) ⇒ Object

Example: {

columns:
  [
    {
      name: id,
      type: INTEGER
    },
    {
      name: type,
      type: STRING
    },
    {
      name: record.child1,
      type: STRING
    },
    {
      name: record.child2,
      type: STRING
    },
values:
  [
    [2,"two","child3","child4"],
    [1,"one","child1","child2"]
  ],
total_rows: 2

}

Returns:

  • Hash result of list table_data



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
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 375

def list_table_data(dataset: nil, table: nil, max_results: 100)
  dataset ||= self.dataset
  table ||= self.table

  begin
    logger.info  { "list_table_data(#{project}, #{dataset}, #{table}, max_results: #{max_results})" }
    response = client.list_table_data(project, dataset, table, max_results: max_results)
  rescue Google::Apis::ServerError, Google::Apis::ClientError, Google::Apis::AuthorizationError => e
    if e.status_code == 404 # not found
      raise NotFoundError, "Table #{project}:#{dataset}.#{table} is not found"
    end

    response = {status_code: e.status_code, message: e.message, error_class: e.class}
    logger.error  { "list_table_data(#{project}, #{dataset}, #{table}, max_results: #{max_results})" }
    raise Error, "Failed to list table_data #{project}:#{dataset}.#{table}, response:#{response}"
  end

  columns = existing_columns
  flattened_columns = Schema.new(columns).flattened_columns.map do |name, column|
    {name: name}.merge!(column)
  end
  if rows = response.to_h[:rows]
    values = TableData.new(columns, rows).values
  end

  {
    total_rows: response.total_rows,
    columns: flattened_columns,
    values: values,
    responses: {
      list_table_data: response,
    }
  }
end

#list_tables(dataset: nil, max_results: 999999) ⇒ Object



249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 249

def list_tables(dataset: nil, max_results: 999999)
  dataset ||= self.dataset

  tables = []
  begin
    logger.info { "List tables... #{project}:#{dataset}" }
    response = client.list_tables(project, dataset, max_results: max_results)
    while true
      _tables = (response.tables || []).map { |t| t.table_reference.table_id.to_s }
      tables.concat(_tables)
      if next_page_token = response.next_page_token
        response = client.list_tables(project, dataset, page_token: next_page_token, max_results: max_results)
      else
        break
      end
    end
  rescue Google::Apis::ServerError, Google::Apis::ClientError, Google::Apis::AuthorizationError => e
    if e.status_code == 404 && /Not found:/ =~ e.message
      raise NotFoundError, "Dataset #{project}:#{dataset} is not found"
    end

    response = {status_code: e.status_code, message: e.message, error_class: e.class}
    logger.error { "list_tables(#{project}, #{dataset}), response:#{response}" }
    raise Error, "failed to list tables #{project}:#{dataset}, response:#{response}"
  end

  { tables: tables }
end

#locationObject



842
843
844
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 842

def location
  config[:location]
end

#loggerObject



17
18
19
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 17

def logger
  BigqueryMigration.logger
end

#migrate_partitioned_table(table: nil, schema_file: nil, columns: nil, options: {}) ⇒ Object

creates a table with time_partitioning option this version only uses patch table API (no query job) because querying partitioned table should cost lots



684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 684

def migrate_partitioned_table(table: nil, schema_file: nil, columns: nil, options: {})
  table ||= self.table

  if schema_file.nil? and columns.nil?
    raise ArgumentError, '`schema_file` or `columns` is required'
  end
  if schema_file
    columns = HashUtil.deep_symbolize_keys(JSON.parse(File.read(schema_file)))
  end
  Schema.validate_columns!(columns)

  result = {}
  begin
    get_table
  rescue NotFoundError
    before_columns = []
    result = create_partitioned_table(table: table, columns: columns, options: options)
  else
    before_columns = existing_columns
    add_columns  = Schema.diff_columns(before_columns, columns)
    drop_columns = Schema.diff_columns(columns, before_columns)

    if !drop_columns.empty? || !add_columns.empty?
      Schema.make_nullable!(drop_columns) # drop columns will be NULLABLE columns
      Schema.reverse_merge!(columns, patch_columns = drop_columns)
      Schema.reverse_merge!(patch_columns, patch_columns = add_columns)
      patch_table(table: table, columns: patch_columns)
    end
  end

  after_columns = existing_columns

  if after_columns.empty? and !dry_run?
    raise Error, "after_columns is empty. " \
      "before_columns: #{before_columns}, after_columns: #{after_columns}, columns: #{columns}"
  end

  result.merge!( before_columns: before_columns, after_columns: after_columns )
end

#migrate_table(table: nil, schema_file: nil, columns: nil, backup_dataset: nil, backup_table: nil) ⇒ Object



641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 641

def migrate_table(table: nil, schema_file: nil, columns: nil, backup_dataset: nil, backup_table: nil)
  table ||= self.table
  backup_dataset ||= self.dataset

  if schema_file.nil? and columns.nil?
    raise ArgumentError, '`schema_file` or `columns` is required'
  end
  if schema_file
    columns = HashUtil.deep_symbolize_keys(JSON.parse(File.read(schema_file)))
  end
  Schema.validate_columns!(columns)

  result = {}
  begin
    get_table
  rescue NotFoundError
    before_columns = []
    result = create_table(table: table, columns: columns)
  else
    before_columns = existing_columns
    add_columns  = Schema.diff_columns(before_columns, columns)
    drop_columns = Schema.diff_columns(columns, before_columns)

    if !drop_columns.empty?
      drop_column(table: table, columns: columns,
                  backup_dataset: backup_dataset, backup_table: backup_table)
    elsif !add_columns.empty?
      add_column(table: table, columns: columns)
    end
  end

  after_columns = existing_columns

  if after_columns.empty? and !dry_run?
    raise Error, "after_columns is empty. " \
      "before_columns: #{before_columns}, after_columns: #{after_columns}, columns: #{columns}"
  end

  result.merge!( before_columns: before_columns, after_columns: after_columns )
end

#open_timeout_secObject



824
825
826
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 824

def open_timeout_sec
  @open_timeout_sec ||= ENV['OPEN_TIMEOUT_SEC'] || config.fetch(:open_timeout_sec, nil) || 300
end

#patch_table(dataset: nil, table: nil, columns: nil, add_columns: nil) ⇒ Object Also known as: add_column



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
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 410

def patch_table(dataset: nil, table: nil, columns: nil, add_columns: nil)
  dataset ||= self.dataset
  table ||= self.table

  if columns.nil? and add_columns.nil?
    raise ArgumentError, 'patch_table: `columns` or `add_columns` is required'
  end

  before_columns = existing_columns
  if columns # if already given
    schema = Schema.new(columns)
  else
    schema = Schema.new(add_columns)
    schema.reverse_merge!(before_columns)
  end
  schema.validate_permitted_operations!(before_columns)

  begin
    logger.info { "#{head}Patch table... #{project}:#{dataset}.#{table}" }
    fields = schema.map {|column| HashUtil.deep_symbolize_keys(column) }
    body = {
      schema: {
        fields: fields,
      }
    }
    opts = {}
    logger.debug { "#{head}patch_table(#{project}, #{dataset}, #{table}, #{body}, options: #{opts})" }
    unless dry_run?
      response = client.patch_table(project, dataset, table, body, options: opts)
    end
  rescue Google::Apis::ServerError, Google::Apis::ClientError, Google::Apis::AuthorizationError => e
    if e.status_code == 404 # not found
      raise NotFoundError, "Table #{project}:#{dataset}.#{table} is not found"
    end

    response = {status_code: e.status_code, message: e.message, error_class: e.class}
    logger.error {
      "patch_table(#{project}, #{dataset}, #{table}, #{body}, options: #{opts}), response:#{response}"
    }
    raise Error, "Failed to patch table #{project}:#{dataset}.#{table}, response:#{response}"
  end

  after_columns = existing_columns

  {
    before_columns: before_columns,
    after_columns:  after_columns,
    responses: { patch_table: response },
  }
end

#projectObject



828
829
830
831
832
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 828

def project
  @project ||= ENV['GOOGLE_PROJECT'] || config.fetch(:project, nil) || credentials[:project_id]
  @project ||= credentials[:client_email].chomp('.iam.gserviceaccount.com').split('@').last if credentials[:client_email]
  @project ||= project_default || raise(ConfigError, '`project` is required.')
end

#project_defaultObject



795
796
797
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 795

def project_default
  (config_default[:core] || {})[:project]
end

#purge_tables(dataset: nil, table_prefix:, suffix_format:, purge_before:, timezone: nil) ⇒ Object



278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 278

def purge_tables(dataset: nil, table_prefix: , suffix_format: , purge_before: , timezone: nil)
  dataset ||= self.dataset
  timezone ||= Time.now.strftime('%z')

  before_tables = list_tables[:tables]

  purge_before_t = TimeWithZone.strptime_with_zone(purge_before, suffix_format, timezone)
  tables = before_tables.select do |tbl|
    suffix = tbl.gsub(table_prefix, '')
    begin
      suffix_t = TimeWithZone.strptime_with_zone(suffix, suffix_format, timezone)
    rescue
      next
    end
    # skip if different from the suffix_format
    next if suffix_t.strftime(suffix_format) != suffix
    suffix_t <= purge_before_t
  end

  tables.each do |_table|
    delete_table(table: _table)
    # If you make more than 100 requests per second, throttling might occur.
    # See https://cloud.google.com/bigquery/quota-policy#apirequests
    sleep 1
  end

  { delete_tables: tables }
end

#read_timeout_secObject



820
821
822
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 820

def read_timeout_sec
  @read_timeout_sec ||= ENV['READ_TIMEOUT_SEC'] || config.fetch(:read_timeout_sec, nil) || timeout_sec || 300
end

#retriesObject



807
808
809
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 807

def retries
  @retries ||= ENV['RETRIES'] || config.fetch(:retries, nil) || 5
end

#send_timeout_secObject



816
817
818
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 816

def send_timeout_sec
  @send_timeout_sec ||= ENV['SEND_TIMEOUT_SEC'] || config.fetch(:send_timeout_sec, nil) || 60
end

#service_accountObject



803
804
805
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 803

def 
  @service_account ||= ENV['GOOGLE_SERVICE_ACCOUNT'] || config.fetch(:service_account, nil) || credentials[:client_email] || 
end

#service_account_defaultObject



791
792
793
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 791

def 
  (config_default[:core] || {})[:account]
end

#support_location_keyword?Boolean

the location keyword arguments are available in google-api-client v0.19.6 or later

Returns:

  • (Boolean)


725
726
727
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 725

def support_location_keyword?
  @support_location_keyword ||= client.method(:get_job).parameters.include?([:key, :location])
end

#tableObject



838
839
840
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 838

def table
  @table  ||= config[:table]   || raise(ConfigError, '`table` is required.')
end

#timeout_secObject

For google-api-client < 0.11.0. Deprecated



812
813
814
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 812

def timeout_sec
  @timeout_sec ||= ENV['TIMEOUT_SEC'] || config.fetch(:timeout_sec, nil)
end

#wait_load(kind, response) ⇒ Object



548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 548

def wait_load(kind, response)
  started = Time.now

  wait_interval = self.job_status_polling_interval
  max_polling_time = self.job_status_max_polling_time
  _response = response

  while true
    job_id = _response.job_reference.job_id
    elapsed = Time.now - started
    status = _response.status.state
    if status == "DONE"
      logger.info {
        "#{kind} job completed... " \
        "job_id:[#{job_id}] elapsed_time:#{elapsed.to_f}sec status:[#{status}]"
      }
      break
    elsif elapsed.to_i > max_polling_time
      message = "#{kind} job checking... " \
        "job_id:[#{job_id}] elapsed_time:#{elapsed.to_f}sec status:[TIMEOUT]"
      logger.info { message }
      raise JobTimeoutError.new(message)
    else
      logger.info {
        "#{kind} job checking... " \
        "job_id:[#{job_id}] elapsed_time:#{elapsed.to_f}sec status:[#{status}]"
      }
      sleep wait_interval
      if support_location_keyword?
        _response = client.get_job(project, job_id, location: location)
      else
        _response = client.get_job(project, job_id)
      end
    end
  end

  # cf. http://www.rubydoc.info/github/google/google-api-ruby-client/Google/Apis/BigqueryV2/JobStatus#errors-instance_method
  # `errors` returns Array<Google::Apis::BigqueryV2::ErrorProto> if any error exists.
  # Otherwise, this returns nil.
  if _errors = _response.status.errors
    raise Error, "Failed during waiting a job, get_job(#{project}, #{job_id}), errors:#{_errors.map(&:to_h)}"
  end

  _response
end

#zone_defaultObject



799
800
801
# File 'lib/bigquery_migration/bigquery_wrapper.rb', line 799

def zone_default
  (config_default[:compute] || {})[:zone]
end