google-cloud-bigquery 1.19.0 → 1.22.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/CHANGELOG.md +42 -0
- data/TROUBLESHOOTING.md +2 -8
- data/lib/google-cloud-bigquery.rb +8 -2
- data/lib/google/cloud/bigquery/convert.rb +3 -1
- data/lib/google/cloud/bigquery/copy_job.rb +15 -6
- data/lib/google/cloud/bigquery/data.rb +12 -0
- data/lib/google/cloud/bigquery/dataset.rb +119 -28
- data/lib/google/cloud/bigquery/external.rb +24 -0
- data/lib/google/cloud/bigquery/extract_job.rb +153 -45
- data/lib/google/cloud/bigquery/job.rb +198 -0
- data/lib/google/cloud/bigquery/load_job.rb +271 -32
- data/lib/google/cloud/bigquery/model.rb +164 -8
- data/lib/google/cloud/bigquery/project.rb +242 -84
- data/lib/google/cloud/bigquery/query_job.rb +277 -6
- data/lib/google/cloud/bigquery/schema.rb +2 -2
- data/lib/google/cloud/bigquery/service.rb +16 -12
- data/lib/google/cloud/bigquery/table.rb +371 -83
- data/lib/google/cloud/bigquery/time.rb +6 -0
- data/lib/google/cloud/bigquery/version.rb +1 -1
- metadata +7 -7
@@ -348,7 +348,71 @@ module Google
|
|
348
348
|
end
|
349
349
|
|
350
350
|
###
|
351
|
-
# Checks if the destination table will be
|
351
|
+
# Checks if the destination table will be range partitioned. See [Creating and using integer range partitioned
|
352
|
+
# tables](https://cloud.google.com/bigquery/docs/creating-integer-range-partitions).
|
353
|
+
#
|
354
|
+
# @return [Boolean] `true` when the table is range partitioned, or `false` otherwise.
|
355
|
+
#
|
356
|
+
# @!group Attributes
|
357
|
+
#
|
358
|
+
def range_partitioning?
|
359
|
+
!@gapi.configuration.load.range_partitioning.nil?
|
360
|
+
end
|
361
|
+
|
362
|
+
###
|
363
|
+
# The field on which the destination table will be range partitioned, if any. The field must be a
|
364
|
+
# top-level `NULLABLE/REQUIRED` field. The only supported type is `INTEGER/INT64`. See
|
365
|
+
# [Creating and using integer range partitioned
|
366
|
+
# tables](https://cloud.google.com/bigquery/docs/creating-integer-range-partitions).
|
367
|
+
#
|
368
|
+
# @return [String, nil] The partition field, if a field was configured, or `nil` if not range partitioned.
|
369
|
+
#
|
370
|
+
# @!group Attributes
|
371
|
+
#
|
372
|
+
def range_partitioning_field
|
373
|
+
@gapi.configuration.load.range_partitioning.field if range_partitioning?
|
374
|
+
end
|
375
|
+
|
376
|
+
###
|
377
|
+
# The start of range partitioning, inclusive. See [Creating and using integer range partitioned
|
378
|
+
# tables](https://cloud.google.com/bigquery/docs/creating-integer-range-partitions).
|
379
|
+
#
|
380
|
+
# @return [Integer, nil] The start of range partitioning, inclusive, or `nil` if not range partitioned.
|
381
|
+
#
|
382
|
+
# @!group Attributes
|
383
|
+
#
|
384
|
+
def range_partitioning_start
|
385
|
+
@gapi.configuration.load.range_partitioning.range.start if range_partitioning?
|
386
|
+
end
|
387
|
+
|
388
|
+
###
|
389
|
+
# The width of each interval. See [Creating and using integer range partitioned
|
390
|
+
# tables](https://cloud.google.com/bigquery/docs/creating-integer-range-partitions).
|
391
|
+
#
|
392
|
+
# @return [Integer, nil] The width of each interval, for data in range partitions, or `nil` if not range
|
393
|
+
# partitioned.
|
394
|
+
#
|
395
|
+
# @!group Attributes
|
396
|
+
#
|
397
|
+
def range_partitioning_interval
|
398
|
+
return nil unless range_partitioning?
|
399
|
+
@gapi.configuration.load.range_partitioning.range.interval
|
400
|
+
end
|
401
|
+
|
402
|
+
###
|
403
|
+
# The end of range partitioning, exclusive. See [Creating and using integer range partitioned
|
404
|
+
# tables](https://cloud.google.com/bigquery/docs/creating-integer-range-partitions).
|
405
|
+
#
|
406
|
+
# @return [Integer, nil] The end of range partitioning, exclusive, or `nil` if not range partitioned.
|
407
|
+
#
|
408
|
+
# @!group Attributes
|
409
|
+
#
|
410
|
+
def range_partitioning_end
|
411
|
+
@gapi.configuration.load.range_partitioning.range.end if range_partitioning?
|
412
|
+
end
|
413
|
+
|
414
|
+
###
|
415
|
+
# Checks if the destination table will be time partitioned. See
|
352
416
|
# [Partitioned Tables](https://cloud.google.com/bigquery/docs/partitioned-tables).
|
353
417
|
#
|
354
418
|
# @return [Boolean, nil] `true` when the table will be time-partitioned,
|
@@ -361,10 +425,10 @@ module Google
|
|
361
425
|
end
|
362
426
|
|
363
427
|
###
|
364
|
-
# The period for which the destination table will be partitioned, if
|
428
|
+
# The period for which the destination table will be time partitioned, if
|
365
429
|
# any. See [Partitioned Tables](https://cloud.google.com/bigquery/docs/partitioned-tables).
|
366
430
|
#
|
367
|
-
# @return [String, nil] The partition type. Currently the only supported
|
431
|
+
# @return [String, nil] The time partition type. Currently the only supported
|
368
432
|
# value is "DAY", or `nil` if not present.
|
369
433
|
#
|
370
434
|
# @!group Attributes
|
@@ -374,13 +438,13 @@ module Google
|
|
374
438
|
end
|
375
439
|
|
376
440
|
###
|
377
|
-
# The field on which the destination table will be partitioned, if any.
|
378
|
-
# If not set, the destination table will be partitioned by pseudo column
|
379
|
-
# `_PARTITIONTIME`; if set, the table will be partitioned by this field.
|
441
|
+
# The field on which the destination table will be time partitioned, if any.
|
442
|
+
# If not set, the destination table will be time partitioned by pseudo column
|
443
|
+
# `_PARTITIONTIME`; if set, the table will be time partitioned by this field.
|
380
444
|
# See [Partitioned Tables](https://cloud.google.com/bigquery/docs/partitioned-tables).
|
381
445
|
#
|
382
|
-
# @return [String, nil] The partition field, if a field was configured.
|
383
|
-
# `nil` if not partitioned or not set (partitioned by pseudo column
|
446
|
+
# @return [String, nil] The time partition field, if a field was configured.
|
447
|
+
# `nil` if not time partitioned or not set (partitioned by pseudo column
|
384
448
|
# '_PARTITIONTIME').
|
385
449
|
#
|
386
450
|
# @!group Attributes
|
@@ -390,12 +454,12 @@ module Google
|
|
390
454
|
end
|
391
455
|
|
392
456
|
###
|
393
|
-
# The expiration for the destination table partitions, if any, in
|
457
|
+
# The expiration for the destination table time partitions, if any, in
|
394
458
|
# seconds. See [Partitioned
|
395
459
|
# Tables](https://cloud.google.com/bigquery/docs/partitioned-tables).
|
396
460
|
#
|
397
461
|
# @return [Integer, nil] The expiration time, in seconds, for data in
|
398
|
-
# partitions, or `nil` if not present.
|
462
|
+
# time partitions, or `nil` if not present.
|
399
463
|
#
|
400
464
|
# @!group Attributes
|
401
465
|
#
|
@@ -408,11 +472,11 @@ module Google
|
|
408
472
|
|
409
473
|
###
|
410
474
|
# If set to true, queries over the destination table will require a
|
411
|
-
# partition filter that can be used for partition elimination to be
|
475
|
+
# time partition filter that can be used for partition elimination to be
|
412
476
|
# specified. See [Partitioned
|
413
477
|
# Tables](https://cloud.google.com/bigquery/docs/partitioned-tables).
|
414
478
|
#
|
415
|
-
# @return [Boolean] `true` when a partition filter will be required,
|
479
|
+
# @return [Boolean] `true` when a time partition filter will be required,
|
416
480
|
# or `false` otherwise.
|
417
481
|
#
|
418
482
|
# @!group Attributes
|
@@ -1239,12 +1303,21 @@ module Google
|
|
1239
1303
|
# Sets the labels to use for the load job.
|
1240
1304
|
#
|
1241
1305
|
# @param [Hash] val A hash of user-provided labels associated with
|
1242
|
-
# the job. You can use these to organize and group your jobs.
|
1243
|
-
#
|
1244
|
-
#
|
1245
|
-
#
|
1246
|
-
#
|
1247
|
-
#
|
1306
|
+
# the job. You can use these to organize and group your jobs.
|
1307
|
+
#
|
1308
|
+
# The labels applied to a resource must meet the following requirements:
|
1309
|
+
#
|
1310
|
+
# * Each resource can have multiple labels, up to a maximum of 64.
|
1311
|
+
# * Each label must be a key-value pair.
|
1312
|
+
# * Keys have a minimum length of 1 character and a maximum length of
|
1313
|
+
# 63 characters, and cannot be empty. Values can be empty, and have
|
1314
|
+
# a maximum length of 63 characters.
|
1315
|
+
# * Keys and values can contain only lowercase letters, numeric characters,
|
1316
|
+
# underscores, and dashes. All characters must use UTF-8 encoding, and
|
1317
|
+
# international characters are allowed.
|
1318
|
+
# * The key portion of a label must be unique. However, you can use the
|
1319
|
+
# same key with multiple resources.
|
1320
|
+
# * Keys must start with a lowercase letter or international character.
|
1248
1321
|
#
|
1249
1322
|
# @!group Attributes
|
1250
1323
|
#
|
@@ -1253,14 +1326,180 @@ module Google
|
|
1253
1326
|
end
|
1254
1327
|
|
1255
1328
|
##
|
1256
|
-
# Sets the
|
1329
|
+
# Sets the field on which to range partition the table. See [Creating and using integer range partitioned
|
1330
|
+
# tables](https://cloud.google.com/bigquery/docs/creating-integer-range-partitions).
|
1331
|
+
#
|
1332
|
+
# See {#range_partitioning_start=}, {#range_partitioning_interval=} and {#range_partitioning_end=}.
|
1333
|
+
#
|
1334
|
+
# You can only set range partitioning when creating a table. BigQuery does not allow you to change
|
1335
|
+
# partitioning on an existing table.
|
1336
|
+
#
|
1337
|
+
# @param [String] field The range partition field. the destination table is partitioned by this
|
1338
|
+
# field. The field must be a top-level `NULLABLE/REQUIRED` field. The only supported
|
1339
|
+
# type is `INTEGER/INT64`.
|
1340
|
+
#
|
1341
|
+
# @example
|
1342
|
+
# require "google/cloud/bigquery"
|
1343
|
+
#
|
1344
|
+
# bigquery = Google::Cloud::Bigquery.new
|
1345
|
+
# dataset = bigquery.dataset "my_dataset"
|
1346
|
+
#
|
1347
|
+
# gs_url = "gs://my-bucket/file-name.csv"
|
1348
|
+
# load_job = dataset.load_job "my_new_table", gs_url do |job|
|
1349
|
+
# job.schema do |schema|
|
1350
|
+
# schema.integer "my_table_id", mode: :required
|
1351
|
+
# schema.string "my_table_data", mode: :required
|
1352
|
+
# end
|
1353
|
+
# job.range_partitioning_field = "my_table_id"
|
1354
|
+
# job.range_partitioning_start = 0
|
1355
|
+
# job.range_partitioning_interval = 10
|
1356
|
+
# job.range_partitioning_end = 100
|
1357
|
+
# end
|
1358
|
+
#
|
1359
|
+
# load_job.wait_until_done!
|
1360
|
+
# load_job.done? #=> true
|
1361
|
+
#
|
1362
|
+
# @!group Attributes
|
1363
|
+
#
|
1364
|
+
def range_partitioning_field= field
|
1365
|
+
@gapi.configuration.load.range_partitioning ||= Google::Apis::BigqueryV2::RangePartitioning.new(
|
1366
|
+
range: Google::Apis::BigqueryV2::RangePartitioning::Range.new
|
1367
|
+
)
|
1368
|
+
@gapi.configuration.load.range_partitioning.field = field
|
1369
|
+
end
|
1370
|
+
|
1371
|
+
##
|
1372
|
+
# Sets the start of range partitioning, inclusive, for the destination table. See [Creating and using integer
|
1373
|
+
# range partitioned tables](https://cloud.google.com/bigquery/docs/creating-integer-range-partitions).
|
1374
|
+
#
|
1375
|
+
# You can only set range partitioning when creating a table. BigQuery does not allow you to change
|
1376
|
+
# partitioning on an existing table.
|
1377
|
+
#
|
1378
|
+
# See {#range_partitioning_field=}, {#range_partitioning_interval=} and {#range_partitioning_end=}.
|
1379
|
+
#
|
1380
|
+
# @param [Integer] range_start The start of range partitioning, inclusive.
|
1381
|
+
#
|
1382
|
+
# @example
|
1383
|
+
# require "google/cloud/bigquery"
|
1384
|
+
#
|
1385
|
+
# bigquery = Google::Cloud::Bigquery.new
|
1386
|
+
# dataset = bigquery.dataset "my_dataset"
|
1387
|
+
#
|
1388
|
+
# gs_url = "gs://my-bucket/file-name.csv"
|
1389
|
+
# load_job = dataset.load_job "my_new_table", gs_url do |job|
|
1390
|
+
# job.schema do |schema|
|
1391
|
+
# schema.integer "my_table_id", mode: :required
|
1392
|
+
# schema.string "my_table_data", mode: :required
|
1393
|
+
# end
|
1394
|
+
# job.range_partitioning_field = "my_table_id"
|
1395
|
+
# job.range_partitioning_start = 0
|
1396
|
+
# job.range_partitioning_interval = 10
|
1397
|
+
# job.range_partitioning_end = 100
|
1398
|
+
# end
|
1399
|
+
#
|
1400
|
+
# load_job.wait_until_done!
|
1401
|
+
# load_job.done? #=> true
|
1402
|
+
#
|
1403
|
+
# @!group Attributes
|
1404
|
+
#
|
1405
|
+
def range_partitioning_start= range_start
|
1406
|
+
@gapi.configuration.load.range_partitioning ||= Google::Apis::BigqueryV2::RangePartitioning.new(
|
1407
|
+
range: Google::Apis::BigqueryV2::RangePartitioning::Range.new
|
1408
|
+
)
|
1409
|
+
@gapi.configuration.load.range_partitioning.range.start = range_start
|
1410
|
+
end
|
1411
|
+
|
1412
|
+
##
|
1413
|
+
# Sets width of each interval for data in range partitions. See [Creating and using integer range partitioned
|
1414
|
+
# tables](https://cloud.google.com/bigquery/docs/creating-integer-range-partitions).
|
1415
|
+
#
|
1416
|
+
# You can only set range partitioning when creating a table. BigQuery does not allow you to change
|
1417
|
+
# partitioning on an existing table.
|
1418
|
+
#
|
1419
|
+
# See {#range_partitioning_field=}, {#range_partitioning_start=} and {#range_partitioning_end=}.
|
1420
|
+
#
|
1421
|
+
# @param [Integer] range_interval The width of each interval, for data in partitions.
|
1422
|
+
#
|
1423
|
+
# @example
|
1424
|
+
# require "google/cloud/bigquery"
|
1425
|
+
#
|
1426
|
+
# bigquery = Google::Cloud::Bigquery.new
|
1427
|
+
# dataset = bigquery.dataset "my_dataset"
|
1428
|
+
#
|
1429
|
+
# gs_url = "gs://my-bucket/file-name.csv"
|
1430
|
+
# load_job = dataset.load_job "my_new_table", gs_url do |job|
|
1431
|
+
# job.schema do |schema|
|
1432
|
+
# schema.integer "my_table_id", mode: :required
|
1433
|
+
# schema.string "my_table_data", mode: :required
|
1434
|
+
# end
|
1435
|
+
# job.range_partitioning_field = "my_table_id"
|
1436
|
+
# job.range_partitioning_start = 0
|
1437
|
+
# job.range_partitioning_interval = 10
|
1438
|
+
# job.range_partitioning_end = 100
|
1439
|
+
# end
|
1440
|
+
#
|
1441
|
+
# load_job.wait_until_done!
|
1442
|
+
# load_job.done? #=> true
|
1443
|
+
#
|
1444
|
+
# @!group Attributes
|
1445
|
+
#
|
1446
|
+
def range_partitioning_interval= range_interval
|
1447
|
+
@gapi.configuration.load.range_partitioning ||= Google::Apis::BigqueryV2::RangePartitioning.new(
|
1448
|
+
range: Google::Apis::BigqueryV2::RangePartitioning::Range.new
|
1449
|
+
)
|
1450
|
+
@gapi.configuration.load.range_partitioning.range.interval = range_interval
|
1451
|
+
end
|
1452
|
+
|
1453
|
+
##
|
1454
|
+
# Sets the end of range partitioning, exclusive, for the destination table. See [Creating and using integer
|
1455
|
+
# range partitioned tables](https://cloud.google.com/bigquery/docs/creating-integer-range-partitions).
|
1456
|
+
#
|
1457
|
+
# You can only set range partitioning when creating a table. BigQuery does not allow you to change
|
1458
|
+
# partitioning on an existing table.
|
1459
|
+
#
|
1460
|
+
# See {#range_partitioning_start=}, {#range_partitioning_interval=} and {#range_partitioning_field=}.
|
1461
|
+
#
|
1462
|
+
# @param [Integer] range_end The end of range partitioning, exclusive.
|
1463
|
+
#
|
1464
|
+
# @example
|
1465
|
+
# require "google/cloud/bigquery"
|
1466
|
+
#
|
1467
|
+
# bigquery = Google::Cloud::Bigquery.new
|
1468
|
+
# dataset = bigquery.dataset "my_dataset"
|
1469
|
+
#
|
1470
|
+
# gs_url = "gs://my-bucket/file-name.csv"
|
1471
|
+
# load_job = dataset.load_job "my_new_table", gs_url do |job|
|
1472
|
+
# job.schema do |schema|
|
1473
|
+
# schema.integer "my_table_id", mode: :required
|
1474
|
+
# schema.string "my_table_data", mode: :required
|
1475
|
+
# end
|
1476
|
+
# job.range_partitioning_field = "my_table_id"
|
1477
|
+
# job.range_partitioning_start = 0
|
1478
|
+
# job.range_partitioning_interval = 10
|
1479
|
+
# job.range_partitioning_end = 100
|
1480
|
+
# end
|
1481
|
+
#
|
1482
|
+
# load_job.wait_until_done!
|
1483
|
+
# load_job.done? #=> true
|
1484
|
+
#
|
1485
|
+
# @!group Attributes
|
1486
|
+
#
|
1487
|
+
def range_partitioning_end= range_end
|
1488
|
+
@gapi.configuration.load.range_partitioning ||= Google::Apis::BigqueryV2::RangePartitioning.new(
|
1489
|
+
range: Google::Apis::BigqueryV2::RangePartitioning::Range.new
|
1490
|
+
)
|
1491
|
+
@gapi.configuration.load.range_partitioning.range.end = range_end
|
1492
|
+
end
|
1493
|
+
|
1494
|
+
##
|
1495
|
+
# Sets the time partitioning for the destination table. See [Partitioned
|
1257
1496
|
# Tables](https://cloud.google.com/bigquery/docs/partitioned-tables).
|
1258
1497
|
#
|
1259
|
-
# You can only set the partitioning field while creating a table.
|
1498
|
+
# You can only set the time partitioning field while creating a table.
|
1260
1499
|
# BigQuery does not allow you to change partitioning on an existing
|
1261
1500
|
# table.
|
1262
1501
|
#
|
1263
|
-
# @param [String] type The partition type. Currently the only
|
1502
|
+
# @param [String] type The time partition type. Currently the only
|
1264
1503
|
# supported value is "DAY".
|
1265
1504
|
#
|
1266
1505
|
# @example
|
@@ -1285,20 +1524,20 @@ module Google
|
|
1285
1524
|
end
|
1286
1525
|
|
1287
1526
|
##
|
1288
|
-
# Sets the field on which to partition the destination table. If not
|
1289
|
-
# set, the destination table is partitioned by pseudo column
|
1290
|
-
# `_PARTITIONTIME`; if set, the table is partitioned by this field.
|
1527
|
+
# Sets the field on which to time partition the destination table. If not
|
1528
|
+
# set, the destination table is time partitioned by pseudo column
|
1529
|
+
# `_PARTITIONTIME`; if set, the table is time partitioned by this field.
|
1291
1530
|
# See [Partitioned
|
1292
1531
|
# Tables](https://cloud.google.com/bigquery/docs/partitioned-tables).
|
1293
1532
|
#
|
1294
|
-
# The destination table must also be partitioned. See
|
1533
|
+
# The destination table must also be time partitioned. See
|
1295
1534
|
# {#time_partitioning_type=}.
|
1296
1535
|
#
|
1297
|
-
# You can only set the partitioning field while creating a table.
|
1536
|
+
# You can only set the time partitioning field while creating a table.
|
1298
1537
|
# BigQuery does not allow you to change partitioning on an existing
|
1299
1538
|
# table.
|
1300
1539
|
#
|
1301
|
-
# @param [String] field The partition field. The field must be a
|
1540
|
+
# @param [String] field The time partition field. The field must be a
|
1302
1541
|
# top-level TIMESTAMP or DATE field. Its mode must be NULLABLE or
|
1303
1542
|
# REQUIRED.
|
1304
1543
|
#
|
@@ -1328,15 +1567,15 @@ module Google
|
|
1328
1567
|
end
|
1329
1568
|
|
1330
1569
|
##
|
1331
|
-
# Sets the partition expiration for the destination table. See
|
1570
|
+
# Sets the time partition expiration for the destination table. See
|
1332
1571
|
# [Partitioned
|
1333
1572
|
# Tables](https://cloud.google.com/bigquery/docs/partitioned-tables).
|
1334
1573
|
#
|
1335
|
-
# The destination table must also be partitioned. See
|
1574
|
+
# The destination table must also be time partitioned. See
|
1336
1575
|
# {#time_partitioning_type=}.
|
1337
1576
|
#
|
1338
1577
|
# @param [Integer] expiration An expiration time, in seconds,
|
1339
|
-
# for data in partitions.
|
1578
|
+
# for data in time partitions.
|
1340
1579
|
#
|
1341
1580
|
# @example
|
1342
1581
|
# require "google/cloud/bigquery"
|
@@ -1362,12 +1601,12 @@ module Google
|
|
1362
1601
|
|
1363
1602
|
##
|
1364
1603
|
# If set to true, queries over the destination table will require a
|
1365
|
-
# partition filter that can be used for partition elimination to be
|
1604
|
+
# time partition filter that can be used for time partition elimination to be
|
1366
1605
|
# specified. See [Partitioned
|
1367
1606
|
# Tables](https://cloud.google.com/bigquery/docs/partitioned-tables).
|
1368
1607
|
#
|
1369
1608
|
# @param [Boolean] val Indicates if queries over the destination table
|
1370
|
-
# will require a partition filter. The default value is `false`.
|
1609
|
+
# will require a time partition filter. The default value is `false`.
|
1371
1610
|
#
|
1372
1611
|
# @!group Attributes
|
1373
1612
|
#
|
@@ -341,14 +341,19 @@ module Google
|
|
341
341
|
# the update to comply with ETag-based optimistic concurrency control.
|
342
342
|
#
|
343
343
|
# @param [Hash<String, String>] new_labels A hash containing key/value
|
344
|
-
# pairs.
|
345
|
-
#
|
346
|
-
# *
|
347
|
-
# *
|
348
|
-
#
|
349
|
-
#
|
350
|
-
#
|
351
|
-
# *
|
344
|
+
# pairs. The labels applied to a resource must meet the following requirements:
|
345
|
+
#
|
346
|
+
# * Each resource can have multiple labels, up to a maximum of 64.
|
347
|
+
# * Each label must be a key-value pair.
|
348
|
+
# * Keys have a minimum length of 1 character and a maximum length of
|
349
|
+
# 63 characters, and cannot be empty. Values can be empty, and have
|
350
|
+
# a maximum length of 63 characters.
|
351
|
+
# * Keys and values can contain only lowercase letters, numeric characters,
|
352
|
+
# underscores, and dashes. All characters must use UTF-8 encoding, and
|
353
|
+
# international characters are allowed.
|
354
|
+
# * The key portion of a label must be unique. However, you can use the
|
355
|
+
# same key with multiple resources.
|
356
|
+
# * Keys must start with a lowercase letter or international character.
|
352
357
|
#
|
353
358
|
# @example
|
354
359
|
# require "google/cloud/bigquery"
|
@@ -482,6 +487,146 @@ module Google
|
|
482
487
|
Array @gapi_json[:trainingRuns]
|
483
488
|
end
|
484
489
|
|
490
|
+
##
|
491
|
+
# Exports the model to Google Cloud Storage asynchronously, immediately
|
492
|
+
# returning an {ExtractJob} that can be used to track the progress of the
|
493
|
+
# export job. The caller may poll the service by repeatedly calling
|
494
|
+
# {Job#reload!} and {Job#done?} to detect when the job is done, or
|
495
|
+
# simply block until the job is done by calling #{Job#wait_until_done!}.
|
496
|
+
# See also {#extract}.
|
497
|
+
#
|
498
|
+
# The geographic location for the job ("US", "EU", etc.) can be set via
|
499
|
+
# {ExtractJob::Updater#location=} in a block passed to this method. If
|
500
|
+
# the model is a full resource representation (see {#resource_full?}),
|
501
|
+
# the location of the job will automatically be set to the location of
|
502
|
+
# the model.
|
503
|
+
#
|
504
|
+
# @see https://cloud.google.com/bigquery-ml/docs/exporting-models
|
505
|
+
# Exporting models
|
506
|
+
#
|
507
|
+
# @param [String] extract_url The Google Storage URI to which BigQuery
|
508
|
+
# should extract the model. This value should be end in an object name
|
509
|
+
# prefix, since multiple objects will be exported.
|
510
|
+
# @param [String] format The exported file format. The default value is
|
511
|
+
# `ml_tf_saved_model`.
|
512
|
+
#
|
513
|
+
# The following values are supported:
|
514
|
+
#
|
515
|
+
# * `ml_tf_saved_model` - TensorFlow SavedModel
|
516
|
+
# * `ml_xgboost_booster` - XGBoost Booster
|
517
|
+
# @param [String] job_id A user-defined ID for the extract job. The ID
|
518
|
+
# must contain only letters (a-z, A-Z), numbers (0-9), underscores
|
519
|
+
# (_), or dashes (-). The maximum length is 1,024 characters. If
|
520
|
+
# `job_id` is provided, then `prefix` will not be used.
|
521
|
+
#
|
522
|
+
# See [Generating a job
|
523
|
+
# ID](https://cloud.google.com/bigquery/docs/managing-jobs#generate-jobid).
|
524
|
+
# @param [String] prefix A string, usually human-readable, that will be
|
525
|
+
# prepended to a generated value to produce a unique job ID. For
|
526
|
+
# example, the prefix `daily_import_job_` can be given to generate a
|
527
|
+
# job ID such as `daily_import_job_12vEDtMQ0mbp1Mo5Z7mzAFQJZazh`. The
|
528
|
+
# prefix must contain only letters (a-z, A-Z), numbers (0-9),
|
529
|
+
# underscores (_), or dashes (-). The maximum length of the entire ID
|
530
|
+
# is 1,024 characters. If `job_id` is provided, then `prefix` will not
|
531
|
+
# be used.
|
532
|
+
# @param [Hash] labels A hash of user-provided labels associated with
|
533
|
+
# the job. You can use these to organize and group your jobs.
|
534
|
+
#
|
535
|
+
# The labels applied to a resource must meet the following requirements:
|
536
|
+
#
|
537
|
+
# * Each resource can have multiple labels, up to a maximum of 64.
|
538
|
+
# * Each label must be a key-value pair.
|
539
|
+
# * Keys have a minimum length of 1 character and a maximum length of
|
540
|
+
# 63 characters, and cannot be empty. Values can be empty, and have
|
541
|
+
# a maximum length of 63 characters.
|
542
|
+
# * Keys and values can contain only lowercase letters, numeric characters,
|
543
|
+
# underscores, and dashes. All characters must use UTF-8 encoding, and
|
544
|
+
# international characters are allowed.
|
545
|
+
# * The key portion of a label must be unique. However, you can use the
|
546
|
+
# same key with multiple resources.
|
547
|
+
# * Keys must start with a lowercase letter or international character.
|
548
|
+
#
|
549
|
+
# @yield [job] a job configuration object
|
550
|
+
# @yieldparam [Google::Cloud::Bigquery::ExtractJob::Updater] job a job
|
551
|
+
# configuration object for setting additional options.
|
552
|
+
#
|
553
|
+
# @return [Google::Cloud::Bigquery::ExtractJob]
|
554
|
+
#
|
555
|
+
# @example
|
556
|
+
# require "google/cloud/bigquery"
|
557
|
+
#
|
558
|
+
# bigquery = Google::Cloud::Bigquery.new
|
559
|
+
# dataset = bigquery.dataset "my_dataset"
|
560
|
+
# model = dataset.model "my_model"
|
561
|
+
#
|
562
|
+
# extract_job = model.extract_job "gs://my-bucket/#{model.model_id}"
|
563
|
+
#
|
564
|
+
# extract_job.wait_until_done!
|
565
|
+
# extract_job.done? #=> true
|
566
|
+
#
|
567
|
+
# @!group Data
|
568
|
+
#
|
569
|
+
def extract_job extract_url, format: nil, job_id: nil, prefix: nil, labels: nil
|
570
|
+
ensure_service!
|
571
|
+
options = { format: format, job_id: job_id, prefix: prefix, labels: labels }
|
572
|
+
updater = ExtractJob::Updater.from_options service, model_ref, extract_url, options
|
573
|
+
updater.location = location if location # may be model reference
|
574
|
+
|
575
|
+
yield updater if block_given?
|
576
|
+
|
577
|
+
job_gapi = updater.to_gapi
|
578
|
+
gapi = service.extract_table job_gapi
|
579
|
+
Job.from_gapi gapi, service
|
580
|
+
end
|
581
|
+
|
582
|
+
##
|
583
|
+
# Exports the model to Google Cloud Storage using a synchronous method
|
584
|
+
# that blocks for a response. Timeouts and transient errors are generally
|
585
|
+
# handled as needed to complete the job. See also {#extract_job}.
|
586
|
+
#
|
587
|
+
# The geographic location for the job ("US", "EU", etc.) can be set via
|
588
|
+
# {ExtractJob::Updater#location=} in a block passed to this method. If
|
589
|
+
# the model is a full resource representation (see {#resource_full?}),
|
590
|
+
# the location of the job will automatically be set to the location of
|
591
|
+
# the model.
|
592
|
+
#
|
593
|
+
# @see https://cloud.google.com/bigquery-ml/docs/exporting-models
|
594
|
+
# Exporting models
|
595
|
+
#
|
596
|
+
# @param [String] extract_url The Google Storage URI to which BigQuery
|
597
|
+
# should extract the model. This value should be end in an object name
|
598
|
+
# prefix, since multiple objects will be exported.
|
599
|
+
# @param [String] format The exported file format. The default value is
|
600
|
+
# `ml_tf_saved_model`.
|
601
|
+
#
|
602
|
+
# The following values are supported:
|
603
|
+
#
|
604
|
+
# * `ml_tf_saved_model` - TensorFlow SavedModel
|
605
|
+
# * `ml_xgboost_booster` - XGBoost Booster
|
606
|
+
# @yield [job] a job configuration object
|
607
|
+
# @yieldparam [Google::Cloud::Bigquery::ExtractJob::Updater] job a job
|
608
|
+
# configuration object for setting additional options.
|
609
|
+
#
|
610
|
+
# @return [Boolean] Returns `true` if the extract operation succeeded.
|
611
|
+
#
|
612
|
+
# @example
|
613
|
+
# require "google/cloud/bigquery"
|
614
|
+
#
|
615
|
+
# bigquery = Google::Cloud::Bigquery.new
|
616
|
+
# dataset = bigquery.dataset "my_dataset"
|
617
|
+
# model = dataset.model "my_model"
|
618
|
+
#
|
619
|
+
# model.extract "gs://my-bucket/#{model.model_id}"
|
620
|
+
#
|
621
|
+
# @!group Data
|
622
|
+
#
|
623
|
+
def extract extract_url, format: nil, &block
|
624
|
+
job = extract_job extract_url, format: format, &block
|
625
|
+
job.wait_until_done!
|
626
|
+
ensure_job_succeeded! job
|
627
|
+
true
|
628
|
+
end
|
629
|
+
|
485
630
|
##
|
486
631
|
# Permanently deletes the model.
|
487
632
|
#
|
@@ -734,6 +879,17 @@ module Google
|
|
734
879
|
def ensure_full_data!
|
735
880
|
reload! unless resource_full?
|
736
881
|
end
|
882
|
+
|
883
|
+
def ensure_job_succeeded! job
|
884
|
+
return unless job.failed?
|
885
|
+
begin
|
886
|
+
# raise to activate ruby exception cause handling
|
887
|
+
raise job.gapi_error
|
888
|
+
rescue StandardError => e
|
889
|
+
# wrap Google::Apis::Error with Google::Cloud::Error
|
890
|
+
raise Google::Cloud::Error.from_error(e)
|
891
|
+
end
|
892
|
+
end
|
737
893
|
end
|
738
894
|
end
|
739
895
|
end
|