/
dataflow_v1b3.projects.jobs.workItems.html
1186 lines (1158 loc) 路 132 KB
/
dataflow_v1b3.projects.jobs.workItems.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
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
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
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
138
139
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
169
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
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
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
277
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
306
307
308
309
310
311
312
313
314
315
316
317
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
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
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
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
537
538
539
540
541
542
543
544
545
546
547
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
593
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
640
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
681
682
683
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
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
<html><body>
<style>
body, h1, h2, h3, div, span, p, pre, a {
margin: 0;
padding: 0;
border: 0;
font-weight: inherit;
font-style: inherit;
font-size: 100%;
font-family: inherit;
vertical-align: baseline;
}
body {
font-size: 13px;
padding: 1em;
}
h1 {
font-size: 26px;
margin-bottom: 1em;
}
h2 {
font-size: 24px;
margin-bottom: 1em;
}
h3 {
font-size: 20px;
margin-bottom: 1em;
margin-top: 1em;
}
pre, code {
line-height: 1.5;
font-family: Monaco, 'DejaVu Sans Mono', 'Bitstream Vera Sans Mono', 'Lucida Console', monospace;
}
pre {
margin-top: 0.5em;
}
h1, h2, h3, p {
font-family: Arial, sans serif;
}
h1, h2, h3 {
border-bottom: solid #CCC 1px;
}
.toc_element {
margin-top: 0.5em;
}
.firstline {
margin-left: 2 em;
}
.method {
margin-top: 1em;
border: solid 1px #CCC;
padding: 1em;
background: #EEE;
}
.details {
font-weight: bold;
font-size: 14px;
}
</style>
<h1><a href="dataflow_v1b3.html">Dataflow API</a> . <a href="dataflow_v1b3.projects.html">projects</a> . <a href="dataflow_v1b3.projects.jobs.html">jobs</a> . <a href="dataflow_v1b3.projects.jobs.workItems.html">workItems</a></h1>
<h2>Instance Methods</h2>
<p class="toc_element">
<code><a href="#close">close()</a></code></p>
<p class="firstline">Close httplib2 connections.</p>
<p class="toc_element">
<code><a href="#lease">lease(projectId, jobId, body=None, x__xgafv=None)</a></code></p>
<p class="firstline">Leases a dataflow WorkItem to run.</p>
<p class="toc_element">
<code><a href="#reportStatus">reportStatus(projectId, jobId, body=None, x__xgafv=None)</a></code></p>
<p class="firstline">Reports the status of dataflow WorkItems leased by a worker.</p>
<h3>Method Details</h3>
<div class="method">
<code class="details" id="close">close()</code>
<pre>Close httplib2 connections.</pre>
</div>
<div class="method">
<code class="details" id="lease">lease(projectId, jobId, body=None, x__xgafv=None)</code>
<pre>Leases a dataflow WorkItem to run.
Args:
projectId: string, Identifies the project this worker belongs to. (required)
jobId: string, Identifies the workflow job this worker belongs to. (required)
body: object, The request body.
The object takes the form of:
{ # Request to lease WorkItems.
"currentWorkerTime": "A String", # The current timestamp at the worker.
"location": "A String", # The [regional endpoint] (https://cloud.google.com/dataflow/docs/concepts/regional-endpoints) that contains the WorkItem's job.
"requestedLeaseDuration": "A String", # The initial lease period.
"unifiedWorkerRequest": { # Untranslated bag-of-bytes WorkRequest from UnifiedWorker.
"a_key": "", # Properties of the object. Contains field @type with type URL.
},
"workItemTypes": [ # Filter for WorkItem type.
"A String",
],
"workerCapabilities": [ # Worker capabilities. WorkItems might be limited to workers with specific capabilities.
"A String",
],
"workerId": "A String", # Identifies the worker leasing work -- typically the ID of the virtual machine running the worker.
}
x__xgafv: string, V1 error format.
Allowed values
1 - v1 error format
2 - v2 error format
Returns:
An object of the form:
{ # Response to a request to lease WorkItems.
"unifiedWorkerResponse": { # Untranslated bag-of-bytes WorkResponse for UnifiedWorker.
"a_key": "", # Properties of the object. Contains field @type with type URL.
},
"workItems": [ # A list of the leased WorkItems.
{ # WorkItem represents basic information about a WorkItem to be executed in the cloud.
"configuration": "A String", # Work item-specific configuration as an opaque blob.
"id": "A String", # Identifies this WorkItem.
"initialReportIndex": "A String", # The initial index to use when reporting the status of the WorkItem.
"jobId": "A String", # Identifies the workflow job this WorkItem belongs to.
"leaseExpireTime": "A String", # Time when the lease on this Work will expire.
"mapTask": { # MapTask consists of an ordered set of instructions, each of which describes one particular low-level operation for the worker to perform in order to accomplish the MapTask's WorkItem. Each instruction must appear in the list before any instructions which depends on its output. # Additional information for MapTask WorkItems.
"counterPrefix": "A String", # Counter prefix that can be used to prefix counters. Not currently used in Dataflow.
"instructions": [ # The instructions in the MapTask.
{ # Describes a particular operation comprising a MapTask.
"flatten": { # An instruction that copies its inputs (zero or more) to its (single) output. # Additional information for Flatten instructions.
"inputs": [ # Describes the inputs to the flatten instruction.
{ # An input of an instruction, as a reference to an output of a producer instruction.
"outputNum": 42, # The output index (origin zero) within the producer.
"producerInstructionIndex": 42, # The index (origin zero) of the parallel instruction that produces the output to be consumed by this input. This index is relative to the list of instructions in this input's instruction's containing MapTask.
},
],
},
"name": "A String", # User-provided name of this operation.
"originalName": "A String", # System-defined name for the operation in the original workflow graph.
"outputs": [ # Describes the outputs of the instruction.
{ # An output of an instruction.
"codec": { # The codec to use to encode data being written via this output.
"a_key": "", # Properties of the object.
},
"name": "A String", # The user-provided name of this output.
"onlyCountKeyBytes": True or False, # For system-generated byte and mean byte metrics, certain instructions should only report the key size.
"onlyCountValueBytes": True or False, # For system-generated byte and mean byte metrics, certain instructions should only report the value size.
"originalName": "A String", # System-defined name for this output in the original workflow graph. Outputs that do not contribute to an original instruction do not set this.
"systemName": "A String", # System-defined name of this output. Unique across the workflow.
},
],
"parDo": { # An instruction that does a ParDo operation. Takes one main input and zero or more side inputs, and produces zero or more outputs. Runs user code. # Additional information for ParDo instructions.
"input": { # An input of an instruction, as a reference to an output of a producer instruction. # The input.
"outputNum": 42, # The output index (origin zero) within the producer.
"producerInstructionIndex": 42, # The index (origin zero) of the parallel instruction that produces the output to be consumed by this input. This index is relative to the list of instructions in this input's instruction's containing MapTask.
},
"multiOutputInfos": [ # Information about each of the outputs, if user_fn is a MultiDoFn.
{ # Information about an output of a multi-output DoFn.
"tag": "A String", # The id of the tag the user code will emit to this output by; this should correspond to the tag of some SideInputInfo.
},
],
"numOutputs": 42, # The number of outputs.
"sideInputs": [ # Zero or more side inputs.
{ # Information about a side input of a DoFn or an input of a SeqDoFn.
"kind": { # How to interpret the source element(s) as a side input value.
"a_key": "", # Properties of the object.
},
"sources": [ # The source(s) to read element(s) from to get the value of this side input. If more than one source, then the elements are taken from the sources, in the specified order if order matters. At least one source is required.
{ # A source that records can be read and decoded from.
"baseSpecs": [ # While splitting, sources may specify the produced bundles as differences against another source, in order to save backend-side memory and allow bigger jobs. For details, see SourceSplitRequest. To support this use case, the full set of parameters of the source is logically obtained by taking the latest explicitly specified value of each parameter in the order: base_specs (later items win), spec (overrides anything in base_specs).
{
"a_key": "", # Properties of the object.
},
],
"codec": { # The codec to use to decode data read from the source.
"a_key": "", # Properties of the object.
},
"doesNotNeedSplitting": True or False, # Setting this value to true hints to the framework that the source doesn't need splitting, and using SourceSplitRequest on it would yield SOURCE_SPLIT_OUTCOME_USE_CURRENT. E.g. a file splitter may set this to true when splitting a single file into a set of byte ranges of appropriate size, and set this to false when splitting a filepattern into individual files. However, for efficiency, a file splitter may decide to produce file subranges directly from the filepattern to avoid a splitting round-trip. See SourceSplitRequest for an overview of the splitting process. This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"metadata": { # Metadata about a Source useful for automatically optimizing and tuning the pipeline, etc. # Optionally, metadata for this source can be supplied right away, avoiding a SourceGetMetadataOperation roundtrip (see SourceOperationRequest). This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"estimatedSizeBytes": "A String", # An estimate of the total size (in bytes) of the data that would be read from this source. This estimate is in terms of external storage size, before any decompression or other processing done by the reader.
"infinite": True or False, # Specifies that the size of this source is known to be infinite (this is a streaming source).
"producesSortedKeys": True or False, # Whether this source is known to produce key/value pairs with the (encoded) keys in lexicographically sorted order.
},
"spec": { # The source to read from, plus its parameters.
"a_key": "", # Properties of the object.
},
},
],
"tag": "A String", # The id of the tag the user code will access this side input by; this should correspond to the tag of some MultiOutputInfo.
},
],
"userFn": { # The user function to invoke.
"a_key": "", # Properties of the object.
},
},
"partialGroupByKey": { # An instruction that does a partial group-by-key. One input and one output. # Additional information for PartialGroupByKey instructions.
"input": { # An input of an instruction, as a reference to an output of a producer instruction. # Describes the input to the partial group-by-key instruction.
"outputNum": 42, # The output index (origin zero) within the producer.
"producerInstructionIndex": 42, # The index (origin zero) of the parallel instruction that produces the output to be consumed by this input. This index is relative to the list of instructions in this input's instruction's containing MapTask.
},
"inputElementCodec": { # The codec to use for interpreting an element in the input PTable.
"a_key": "", # Properties of the object.
},
"originalCombineValuesInputStoreName": "A String", # If this instruction includes a combining function this is the name of the intermediate store between the GBK and the CombineValues.
"originalCombineValuesStepName": "A String", # If this instruction includes a combining function, this is the name of the CombineValues instruction lifted into this instruction.
"sideInputs": [ # Zero or more side inputs.
{ # Information about a side input of a DoFn or an input of a SeqDoFn.
"kind": { # How to interpret the source element(s) as a side input value.
"a_key": "", # Properties of the object.
},
"sources": [ # The source(s) to read element(s) from to get the value of this side input. If more than one source, then the elements are taken from the sources, in the specified order if order matters. At least one source is required.
{ # A source that records can be read and decoded from.
"baseSpecs": [ # While splitting, sources may specify the produced bundles as differences against another source, in order to save backend-side memory and allow bigger jobs. For details, see SourceSplitRequest. To support this use case, the full set of parameters of the source is logically obtained by taking the latest explicitly specified value of each parameter in the order: base_specs (later items win), spec (overrides anything in base_specs).
{
"a_key": "", # Properties of the object.
},
],
"codec": { # The codec to use to decode data read from the source.
"a_key": "", # Properties of the object.
},
"doesNotNeedSplitting": True or False, # Setting this value to true hints to the framework that the source doesn't need splitting, and using SourceSplitRequest on it would yield SOURCE_SPLIT_OUTCOME_USE_CURRENT. E.g. a file splitter may set this to true when splitting a single file into a set of byte ranges of appropriate size, and set this to false when splitting a filepattern into individual files. However, for efficiency, a file splitter may decide to produce file subranges directly from the filepattern to avoid a splitting round-trip. See SourceSplitRequest for an overview of the splitting process. This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"metadata": { # Metadata about a Source useful for automatically optimizing and tuning the pipeline, etc. # Optionally, metadata for this source can be supplied right away, avoiding a SourceGetMetadataOperation roundtrip (see SourceOperationRequest). This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"estimatedSizeBytes": "A String", # An estimate of the total size (in bytes) of the data that would be read from this source. This estimate is in terms of external storage size, before any decompression or other processing done by the reader.
"infinite": True or False, # Specifies that the size of this source is known to be infinite (this is a streaming source).
"producesSortedKeys": True or False, # Whether this source is known to produce key/value pairs with the (encoded) keys in lexicographically sorted order.
},
"spec": { # The source to read from, plus its parameters.
"a_key": "", # Properties of the object.
},
},
],
"tag": "A String", # The id of the tag the user code will access this side input by; this should correspond to the tag of some MultiOutputInfo.
},
],
"valueCombiningFn": { # The value combining function to invoke.
"a_key": "", # Properties of the object.
},
},
"read": { # An instruction that reads records. Takes no inputs, produces one output. # Additional information for Read instructions.
"source": { # A source that records can be read and decoded from. # The source to read from.
"baseSpecs": [ # While splitting, sources may specify the produced bundles as differences against another source, in order to save backend-side memory and allow bigger jobs. For details, see SourceSplitRequest. To support this use case, the full set of parameters of the source is logically obtained by taking the latest explicitly specified value of each parameter in the order: base_specs (later items win), spec (overrides anything in base_specs).
{
"a_key": "", # Properties of the object.
},
],
"codec": { # The codec to use to decode data read from the source.
"a_key": "", # Properties of the object.
},
"doesNotNeedSplitting": True or False, # Setting this value to true hints to the framework that the source doesn't need splitting, and using SourceSplitRequest on it would yield SOURCE_SPLIT_OUTCOME_USE_CURRENT. E.g. a file splitter may set this to true when splitting a single file into a set of byte ranges of appropriate size, and set this to false when splitting a filepattern into individual files. However, for efficiency, a file splitter may decide to produce file subranges directly from the filepattern to avoid a splitting round-trip. See SourceSplitRequest for an overview of the splitting process. This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"metadata": { # Metadata about a Source useful for automatically optimizing and tuning the pipeline, etc. # Optionally, metadata for this source can be supplied right away, avoiding a SourceGetMetadataOperation roundtrip (see SourceOperationRequest). This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"estimatedSizeBytes": "A String", # An estimate of the total size (in bytes) of the data that would be read from this source. This estimate is in terms of external storage size, before any decompression or other processing done by the reader.
"infinite": True or False, # Specifies that the size of this source is known to be infinite (this is a streaming source).
"producesSortedKeys": True or False, # Whether this source is known to produce key/value pairs with the (encoded) keys in lexicographically sorted order.
},
"spec": { # The source to read from, plus its parameters.
"a_key": "", # Properties of the object.
},
},
},
"systemName": "A String", # System-defined name of this operation. Unique across the workflow.
"write": { # An instruction that writes records. Takes one input, produces no outputs. # Additional information for Write instructions.
"input": { # An input of an instruction, as a reference to an output of a producer instruction. # The input.
"outputNum": 42, # The output index (origin zero) within the producer.
"producerInstructionIndex": 42, # The index (origin zero) of the parallel instruction that produces the output to be consumed by this input. This index is relative to the list of instructions in this input's instruction's containing MapTask.
},
"sink": { # A sink that records can be encoded and written to. # The sink to write to.
"codec": { # The codec to use to encode data written to the sink.
"a_key": "", # Properties of the object.
},
"spec": { # The sink to write to, plus its parameters.
"a_key": "", # Properties of the object.
},
},
},
},
],
"stageName": "A String", # System-defined name of the stage containing this MapTask. Unique across the workflow.
"systemName": "A String", # System-defined name of this MapTask. Unique across the workflow.
},
"packages": [ # Any required packages that need to be fetched in order to execute this WorkItem.
{ # The packages that must be installed in order for a worker to run the steps of the Cloud Dataflow job that will be assigned to its worker pool. This is the mechanism by which the Cloud Dataflow SDK causes code to be loaded onto the workers. For example, the Cloud Dataflow Java SDK might use this to install jars containing the user's code and all of the various dependencies (libraries, data files, etc.) required in order for that code to run.
"location": "A String", # The resource to read the package from. The supported resource type is: Google Cloud Storage: storage.googleapis.com/{bucket} bucket.storage.googleapis.com/
"name": "A String", # The name of the package.
},
],
"projectId": "A String", # Identifies the cloud project this WorkItem belongs to.
"reportStatusInterval": "A String", # Recommended reporting interval.
"seqMapTask": { # Describes a particular function to invoke. # Additional information for SeqMapTask WorkItems.
"inputs": [ # Information about each of the inputs.
{ # Information about a side input of a DoFn or an input of a SeqDoFn.
"kind": { # How to interpret the source element(s) as a side input value.
"a_key": "", # Properties of the object.
},
"sources": [ # The source(s) to read element(s) from to get the value of this side input. If more than one source, then the elements are taken from the sources, in the specified order if order matters. At least one source is required.
{ # A source that records can be read and decoded from.
"baseSpecs": [ # While splitting, sources may specify the produced bundles as differences against another source, in order to save backend-side memory and allow bigger jobs. For details, see SourceSplitRequest. To support this use case, the full set of parameters of the source is logically obtained by taking the latest explicitly specified value of each parameter in the order: base_specs (later items win), spec (overrides anything in base_specs).
{
"a_key": "", # Properties of the object.
},
],
"codec": { # The codec to use to decode data read from the source.
"a_key": "", # Properties of the object.
},
"doesNotNeedSplitting": True or False, # Setting this value to true hints to the framework that the source doesn't need splitting, and using SourceSplitRequest on it would yield SOURCE_SPLIT_OUTCOME_USE_CURRENT. E.g. a file splitter may set this to true when splitting a single file into a set of byte ranges of appropriate size, and set this to false when splitting a filepattern into individual files. However, for efficiency, a file splitter may decide to produce file subranges directly from the filepattern to avoid a splitting round-trip. See SourceSplitRequest for an overview of the splitting process. This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"metadata": { # Metadata about a Source useful for automatically optimizing and tuning the pipeline, etc. # Optionally, metadata for this source can be supplied right away, avoiding a SourceGetMetadataOperation roundtrip (see SourceOperationRequest). This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"estimatedSizeBytes": "A String", # An estimate of the total size (in bytes) of the data that would be read from this source. This estimate is in terms of external storage size, before any decompression or other processing done by the reader.
"infinite": True or False, # Specifies that the size of this source is known to be infinite (this is a streaming source).
"producesSortedKeys": True or False, # Whether this source is known to produce key/value pairs with the (encoded) keys in lexicographically sorted order.
},
"spec": { # The source to read from, plus its parameters.
"a_key": "", # Properties of the object.
},
},
],
"tag": "A String", # The id of the tag the user code will access this side input by; this should correspond to the tag of some MultiOutputInfo.
},
],
"name": "A String", # The user-provided name of the SeqDo operation.
"outputInfos": [ # Information about each of the outputs.
{ # Information about an output of a SeqMapTask.
"sink": { # A sink that records can be encoded and written to. # The sink to write the output value to.
"codec": { # The codec to use to encode data written to the sink.
"a_key": "", # Properties of the object.
},
"spec": { # The sink to write to, plus its parameters.
"a_key": "", # Properties of the object.
},
},
"tag": "A String", # The id of the TupleTag the user code will tag the output value by.
},
],
"stageName": "A String", # System-defined name of the stage containing the SeqDo operation. Unique across the workflow.
"systemName": "A String", # System-defined name of the SeqDo operation. Unique across the workflow.
"userFn": { # The user function to invoke.
"a_key": "", # Properties of the object.
},
},
"shellTask": { # A task which consists of a shell command for the worker to execute. # Additional information for ShellTask WorkItems.
"command": "A String", # The shell command to run.
"exitCode": 42, # Exit code for the task.
},
"sourceOperationTask": { # A work item that represents the different operations that can be performed on a user-defined Source specification. # Additional information for source operation WorkItems.
"getMetadata": { # A request to compute the SourceMetadata of a Source. # Information about a request to get metadata about a source.
"source": { # A source that records can be read and decoded from. # Specification of the source whose metadata should be computed.
"baseSpecs": [ # While splitting, sources may specify the produced bundles as differences against another source, in order to save backend-side memory and allow bigger jobs. For details, see SourceSplitRequest. To support this use case, the full set of parameters of the source is logically obtained by taking the latest explicitly specified value of each parameter in the order: base_specs (later items win), spec (overrides anything in base_specs).
{
"a_key": "", # Properties of the object.
},
],
"codec": { # The codec to use to decode data read from the source.
"a_key": "", # Properties of the object.
},
"doesNotNeedSplitting": True or False, # Setting this value to true hints to the framework that the source doesn't need splitting, and using SourceSplitRequest on it would yield SOURCE_SPLIT_OUTCOME_USE_CURRENT. E.g. a file splitter may set this to true when splitting a single file into a set of byte ranges of appropriate size, and set this to false when splitting a filepattern into individual files. However, for efficiency, a file splitter may decide to produce file subranges directly from the filepattern to avoid a splitting round-trip. See SourceSplitRequest for an overview of the splitting process. This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"metadata": { # Metadata about a Source useful for automatically optimizing and tuning the pipeline, etc. # Optionally, metadata for this source can be supplied right away, avoiding a SourceGetMetadataOperation roundtrip (see SourceOperationRequest). This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"estimatedSizeBytes": "A String", # An estimate of the total size (in bytes) of the data that would be read from this source. This estimate is in terms of external storage size, before any decompression or other processing done by the reader.
"infinite": True or False, # Specifies that the size of this source is known to be infinite (this is a streaming source).
"producesSortedKeys": True or False, # Whether this source is known to produce key/value pairs with the (encoded) keys in lexicographically sorted order.
},
"spec": { # The source to read from, plus its parameters.
"a_key": "", # Properties of the object.
},
},
},
"name": "A String", # User-provided name of the Read instruction for this source.
"originalName": "A String", # System-defined name for the Read instruction for this source in the original workflow graph.
"split": { # Represents the operation to split a high-level Source specification into bundles (parts for parallel processing). At a high level, splitting of a source into bundles happens as follows: SourceSplitRequest is applied to the source. If it returns SOURCE_SPLIT_OUTCOME_USE_CURRENT, no further splitting happens and the source is used "as is". Otherwise, splitting is applied recursively to each produced DerivedSource. As an optimization, for any Source, if its does_not_need_splitting is true, the framework assumes that splitting this source would return SOURCE_SPLIT_OUTCOME_USE_CURRENT, and doesn't initiate a SourceSplitRequest. This applies both to the initial source being split and to bundles produced from it. # Information about a request to split a source.
"options": { # Hints for splitting a Source into bundles (parts for parallel processing) using SourceSplitRequest. # Hints for tuning the splitting process.
"desiredBundleSizeBytes": "A String", # The source should be split into a set of bundles where the estimated size of each is approximately this many bytes.
"desiredShardSizeBytes": "A String", # DEPRECATED in favor of desired_bundle_size_bytes.
},
"source": { # A source that records can be read and decoded from. # Specification of the source to be split.
"baseSpecs": [ # While splitting, sources may specify the produced bundles as differences against another source, in order to save backend-side memory and allow bigger jobs. For details, see SourceSplitRequest. To support this use case, the full set of parameters of the source is logically obtained by taking the latest explicitly specified value of each parameter in the order: base_specs (later items win), spec (overrides anything in base_specs).
{
"a_key": "", # Properties of the object.
},
],
"codec": { # The codec to use to decode data read from the source.
"a_key": "", # Properties of the object.
},
"doesNotNeedSplitting": True or False, # Setting this value to true hints to the framework that the source doesn't need splitting, and using SourceSplitRequest on it would yield SOURCE_SPLIT_OUTCOME_USE_CURRENT. E.g. a file splitter may set this to true when splitting a single file into a set of byte ranges of appropriate size, and set this to false when splitting a filepattern into individual files. However, for efficiency, a file splitter may decide to produce file subranges directly from the filepattern to avoid a splitting round-trip. See SourceSplitRequest for an overview of the splitting process. This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"metadata": { # Metadata about a Source useful for automatically optimizing and tuning the pipeline, etc. # Optionally, metadata for this source can be supplied right away, avoiding a SourceGetMetadataOperation roundtrip (see SourceOperationRequest). This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"estimatedSizeBytes": "A String", # An estimate of the total size (in bytes) of the data that would be read from this source. This estimate is in terms of external storage size, before any decompression or other processing done by the reader.
"infinite": True or False, # Specifies that the size of this source is known to be infinite (this is a streaming source).
"producesSortedKeys": True or False, # Whether this source is known to produce key/value pairs with the (encoded) keys in lexicographically sorted order.
},
"spec": { # The source to read from, plus its parameters.
"a_key": "", # Properties of the object.
},
},
},
"stageName": "A String", # System-defined name of the stage containing the source operation. Unique across the workflow.
"systemName": "A String", # System-defined name of the Read instruction for this source. Unique across the workflow.
},
"streamingComputationTask": { # A task which describes what action should be performed for the specified streaming computation ranges. # Additional information for StreamingComputationTask WorkItems.
"computationRanges": [ # Contains ranges of a streaming computation this task should apply to.
{ # Describes full or partial data disk assignment information of the computation ranges.
"computationId": "A String", # The ID of the computation.
"rangeAssignments": [ # Data disk assignments for ranges from this computation.
{ # Data disk assignment information for a specific key-range of a sharded computation. Currently we only support UTF-8 character splits to simplify encoding into JSON.
"dataDisk": "A String", # The name of the data disk where data for this range is stored. This name is local to the Google Cloud Platform project and uniquely identifies the disk within that project, for example "myproject-1014-104817-4c2-harness-0-disk-1".
"end": "A String", # The end (exclusive) of the key range.
"start": "A String", # The start (inclusive) of the key range.
},
],
},
],
"dataDisks": [ # Describes the set of data disks this task should apply to.
{ # Describes mounted data disk.
"dataDisk": "A String", # The name of the data disk. This name is local to the Google Cloud Platform project and uniquely identifies the disk within that project, for example "myproject-1014-104817-4c2-harness-0-disk-1".
},
],
"taskType": "A String", # A type of streaming computation task.
},
"streamingConfigTask": { # A task that carries configuration information for streaming computations. # Additional information for StreamingConfigTask WorkItems.
"commitStreamChunkSizeBytes": "A String", # Chunk size for commit streams from the harness to windmill.
"getDataStreamChunkSizeBytes": "A String", # Chunk size for get data streams from the harness to windmill.
"maxWorkItemCommitBytes": "A String", # Maximum size for work item commit supported windmill storage layer.
"streamingComputationConfigs": [ # Set of computation configuration information.
{ # Configuration information for a single streaming computation.
"computationId": "A String", # Unique identifier for this computation.
"instructions": [ # Instructions that comprise the computation.
{ # Describes a particular operation comprising a MapTask.
"flatten": { # An instruction that copies its inputs (zero or more) to its (single) output. # Additional information for Flatten instructions.
"inputs": [ # Describes the inputs to the flatten instruction.
{ # An input of an instruction, as a reference to an output of a producer instruction.
"outputNum": 42, # The output index (origin zero) within the producer.
"producerInstructionIndex": 42, # The index (origin zero) of the parallel instruction that produces the output to be consumed by this input. This index is relative to the list of instructions in this input's instruction's containing MapTask.
},
],
},
"name": "A String", # User-provided name of this operation.
"originalName": "A String", # System-defined name for the operation in the original workflow graph.
"outputs": [ # Describes the outputs of the instruction.
{ # An output of an instruction.
"codec": { # The codec to use to encode data being written via this output.
"a_key": "", # Properties of the object.
},
"name": "A String", # The user-provided name of this output.
"onlyCountKeyBytes": True or False, # For system-generated byte and mean byte metrics, certain instructions should only report the key size.
"onlyCountValueBytes": True or False, # For system-generated byte and mean byte metrics, certain instructions should only report the value size.
"originalName": "A String", # System-defined name for this output in the original workflow graph. Outputs that do not contribute to an original instruction do not set this.
"systemName": "A String", # System-defined name of this output. Unique across the workflow.
},
],
"parDo": { # An instruction that does a ParDo operation. Takes one main input and zero or more side inputs, and produces zero or more outputs. Runs user code. # Additional information for ParDo instructions.
"input": { # An input of an instruction, as a reference to an output of a producer instruction. # The input.
"outputNum": 42, # The output index (origin zero) within the producer.
"producerInstructionIndex": 42, # The index (origin zero) of the parallel instruction that produces the output to be consumed by this input. This index is relative to the list of instructions in this input's instruction's containing MapTask.
},
"multiOutputInfos": [ # Information about each of the outputs, if user_fn is a MultiDoFn.
{ # Information about an output of a multi-output DoFn.
"tag": "A String", # The id of the tag the user code will emit to this output by; this should correspond to the tag of some SideInputInfo.
},
],
"numOutputs": 42, # The number of outputs.
"sideInputs": [ # Zero or more side inputs.
{ # Information about a side input of a DoFn or an input of a SeqDoFn.
"kind": { # How to interpret the source element(s) as a side input value.
"a_key": "", # Properties of the object.
},
"sources": [ # The source(s) to read element(s) from to get the value of this side input. If more than one source, then the elements are taken from the sources, in the specified order if order matters. At least one source is required.
{ # A source that records can be read and decoded from.
"baseSpecs": [ # While splitting, sources may specify the produced bundles as differences against another source, in order to save backend-side memory and allow bigger jobs. For details, see SourceSplitRequest. To support this use case, the full set of parameters of the source is logically obtained by taking the latest explicitly specified value of each parameter in the order: base_specs (later items win), spec (overrides anything in base_specs).
{
"a_key": "", # Properties of the object.
},
],
"codec": { # The codec to use to decode data read from the source.
"a_key": "", # Properties of the object.
},
"doesNotNeedSplitting": True or False, # Setting this value to true hints to the framework that the source doesn't need splitting, and using SourceSplitRequest on it would yield SOURCE_SPLIT_OUTCOME_USE_CURRENT. E.g. a file splitter may set this to true when splitting a single file into a set of byte ranges of appropriate size, and set this to false when splitting a filepattern into individual files. However, for efficiency, a file splitter may decide to produce file subranges directly from the filepattern to avoid a splitting round-trip. See SourceSplitRequest for an overview of the splitting process. This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"metadata": { # Metadata about a Source useful for automatically optimizing and tuning the pipeline, etc. # Optionally, metadata for this source can be supplied right away, avoiding a SourceGetMetadataOperation roundtrip (see SourceOperationRequest). This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"estimatedSizeBytes": "A String", # An estimate of the total size (in bytes) of the data that would be read from this source. This estimate is in terms of external storage size, before any decompression or other processing done by the reader.
"infinite": True or False, # Specifies that the size of this source is known to be infinite (this is a streaming source).
"producesSortedKeys": True or False, # Whether this source is known to produce key/value pairs with the (encoded) keys in lexicographically sorted order.
},
"spec": { # The source to read from, plus its parameters.
"a_key": "", # Properties of the object.
},
},
],
"tag": "A String", # The id of the tag the user code will access this side input by; this should correspond to the tag of some MultiOutputInfo.
},
],
"userFn": { # The user function to invoke.
"a_key": "", # Properties of the object.
},
},
"partialGroupByKey": { # An instruction that does a partial group-by-key. One input and one output. # Additional information for PartialGroupByKey instructions.
"input": { # An input of an instruction, as a reference to an output of a producer instruction. # Describes the input to the partial group-by-key instruction.
"outputNum": 42, # The output index (origin zero) within the producer.
"producerInstructionIndex": 42, # The index (origin zero) of the parallel instruction that produces the output to be consumed by this input. This index is relative to the list of instructions in this input's instruction's containing MapTask.
},
"inputElementCodec": { # The codec to use for interpreting an element in the input PTable.
"a_key": "", # Properties of the object.
},
"originalCombineValuesInputStoreName": "A String", # If this instruction includes a combining function this is the name of the intermediate store between the GBK and the CombineValues.
"originalCombineValuesStepName": "A String", # If this instruction includes a combining function, this is the name of the CombineValues instruction lifted into this instruction.
"sideInputs": [ # Zero or more side inputs.
{ # Information about a side input of a DoFn or an input of a SeqDoFn.
"kind": { # How to interpret the source element(s) as a side input value.
"a_key": "", # Properties of the object.
},
"sources": [ # The source(s) to read element(s) from to get the value of this side input. If more than one source, then the elements are taken from the sources, in the specified order if order matters. At least one source is required.
{ # A source that records can be read and decoded from.
"baseSpecs": [ # While splitting, sources may specify the produced bundles as differences against another source, in order to save backend-side memory and allow bigger jobs. For details, see SourceSplitRequest. To support this use case, the full set of parameters of the source is logically obtained by taking the latest explicitly specified value of each parameter in the order: base_specs (later items win), spec (overrides anything in base_specs).
{
"a_key": "", # Properties of the object.
},
],
"codec": { # The codec to use to decode data read from the source.
"a_key": "", # Properties of the object.
},
"doesNotNeedSplitting": True or False, # Setting this value to true hints to the framework that the source doesn't need splitting, and using SourceSplitRequest on it would yield SOURCE_SPLIT_OUTCOME_USE_CURRENT. E.g. a file splitter may set this to true when splitting a single file into a set of byte ranges of appropriate size, and set this to false when splitting a filepattern into individual files. However, for efficiency, a file splitter may decide to produce file subranges directly from the filepattern to avoid a splitting round-trip. See SourceSplitRequest for an overview of the splitting process. This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"metadata": { # Metadata about a Source useful for automatically optimizing and tuning the pipeline, etc. # Optionally, metadata for this source can be supplied right away, avoiding a SourceGetMetadataOperation roundtrip (see SourceOperationRequest). This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"estimatedSizeBytes": "A String", # An estimate of the total size (in bytes) of the data that would be read from this source. This estimate is in terms of external storage size, before any decompression or other processing done by the reader.
"infinite": True or False, # Specifies that the size of this source is known to be infinite (this is a streaming source).
"producesSortedKeys": True or False, # Whether this source is known to produce key/value pairs with the (encoded) keys in lexicographically sorted order.
},
"spec": { # The source to read from, plus its parameters.
"a_key": "", # Properties of the object.
},
},
],
"tag": "A String", # The id of the tag the user code will access this side input by; this should correspond to the tag of some MultiOutputInfo.
},
],
"valueCombiningFn": { # The value combining function to invoke.
"a_key": "", # Properties of the object.
},
},
"read": { # An instruction that reads records. Takes no inputs, produces one output. # Additional information for Read instructions.
"source": { # A source that records can be read and decoded from. # The source to read from.
"baseSpecs": [ # While splitting, sources may specify the produced bundles as differences against another source, in order to save backend-side memory and allow bigger jobs. For details, see SourceSplitRequest. To support this use case, the full set of parameters of the source is logically obtained by taking the latest explicitly specified value of each parameter in the order: base_specs (later items win), spec (overrides anything in base_specs).
{
"a_key": "", # Properties of the object.
},
],
"codec": { # The codec to use to decode data read from the source.
"a_key": "", # Properties of the object.
},
"doesNotNeedSplitting": True or False, # Setting this value to true hints to the framework that the source doesn't need splitting, and using SourceSplitRequest on it would yield SOURCE_SPLIT_OUTCOME_USE_CURRENT. E.g. a file splitter may set this to true when splitting a single file into a set of byte ranges of appropriate size, and set this to false when splitting a filepattern into individual files. However, for efficiency, a file splitter may decide to produce file subranges directly from the filepattern to avoid a splitting round-trip. See SourceSplitRequest for an overview of the splitting process. This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"metadata": { # Metadata about a Source useful for automatically optimizing and tuning the pipeline, etc. # Optionally, metadata for this source can be supplied right away, avoiding a SourceGetMetadataOperation roundtrip (see SourceOperationRequest). This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"estimatedSizeBytes": "A String", # An estimate of the total size (in bytes) of the data that would be read from this source. This estimate is in terms of external storage size, before any decompression or other processing done by the reader.
"infinite": True or False, # Specifies that the size of this source is known to be infinite (this is a streaming source).
"producesSortedKeys": True or False, # Whether this source is known to produce key/value pairs with the (encoded) keys in lexicographically sorted order.
},
"spec": { # The source to read from, plus its parameters.
"a_key": "", # Properties of the object.
},
},
},
"systemName": "A String", # System-defined name of this operation. Unique across the workflow.
"write": { # An instruction that writes records. Takes one input, produces no outputs. # Additional information for Write instructions.
"input": { # An input of an instruction, as a reference to an output of a producer instruction. # The input.
"outputNum": 42, # The output index (origin zero) within the producer.
"producerInstructionIndex": 42, # The index (origin zero) of the parallel instruction that produces the output to be consumed by this input. This index is relative to the list of instructions in this input's instruction's containing MapTask.
},
"sink": { # A sink that records can be encoded and written to. # The sink to write to.
"codec": { # The codec to use to encode data written to the sink.
"a_key": "", # Properties of the object.
},
"spec": { # The sink to write to, plus its parameters.
"a_key": "", # Properties of the object.
},
},
},
},
],
"stageName": "A String", # Stage name of this computation.
"systemName": "A String", # System defined name for this computation.
"transformUserNameToStateFamily": { # Map from user name of stateful transforms in this stage to their state family.
"a_key": "A String",
},
},
],
"userStepToStateFamilyNameMap": { # Map from user step names to state families.
"a_key": "A String",
},
"windmillServiceEndpoint": "A String", # If present, the worker must use this endpoint to communicate with Windmill Service dispatchers, otherwise the worker must continue to use whatever endpoint it had been using.
"windmillServicePort": "A String", # If present, the worker must use this port to communicate with Windmill Service dispatchers. Only applicable when windmill_service_endpoint is specified.
},
"streamingSetupTask": { # A task which initializes part of a streaming Dataflow job. # Additional information for StreamingSetupTask WorkItems.
"drain": True or False, # The user has requested drain.
"receiveWorkPort": 42, # The TCP port on which the worker should listen for messages from other streaming computation workers.
"snapshotConfig": { # Streaming appliance snapshot configuration. # Configures streaming appliance snapshot.
"importStateEndpoint": "A String", # Indicates which endpoint is used to import appliance state.
"snapshotId": "A String", # If set, indicates the snapshot id for the snapshot being performed.
},
"streamingComputationTopology": { # Global topology of the streaming Dataflow job, including all computations and their sharded locations. # The global topology of the streaming Dataflow job.
"computations": [ # The computations associated with a streaming Dataflow job.
{ # All configuration data for a particular Computation.
"computationId": "A String", # The ID of the computation.
"inputs": [ # The inputs to the computation.
{ # Describes a stream of data, either as input to be processed or as output of a streaming Dataflow job.
"customSourceLocation": { # Identifies the location of a custom souce. # The stream is a custom source.
"stateful": True or False, # Whether this source is stateful.
},
"pubsubLocation": { # Identifies a pubsub location to use for transferring data into or out of a streaming Dataflow job. # The stream is a pubsub stream.
"dropLateData": True or False, # Indicates whether the pipeline allows late-arriving data.
"dynamicDestinations": True or False, # If true, then this location represents dynamic topics.
"idLabel": "A String", # If set, contains a pubsub label from which to extract record ids. If left empty, record deduplication will be strictly best effort.
"subscription": "A String", # A pubsub subscription, in the form of "pubsub.googleapis.com/subscriptions//"
"timestampLabel": "A String", # If set, contains a pubsub label from which to extract record timestamps. If left empty, record timestamps will be generated upon arrival.
"topic": "A String", # A pubsub topic, in the form of "pubsub.googleapis.com/topics//"
"trackingSubscription": "A String", # If set, specifies the pubsub subscription that will be used for tracking custom time timestamps for watermark estimation.
"withAttributes": True or False, # If true, then the client has requested to get pubsub attributes.
},
"sideInputLocation": { # Identifies the location of a streaming side input. # The stream is a streaming side input.
"stateFamily": "A String", # Identifies the state family where this side input is stored.
"tag": "A String", # Identifies the particular side input within the streaming Dataflow job.
},
"streamingStageLocation": { # Identifies the location of a streaming computation stage, for stage-to-stage communication. # The stream is part of another computation within the current streaming Dataflow job.
"streamId": "A String", # Identifies the particular stream within the streaming Dataflow job.
},
},
],
"keyRanges": [ # The key ranges processed by the computation.
{ # Location information for a specific key-range of a sharded computation. Currently we only support UTF-8 character splits to simplify encoding into JSON.
"dataDisk": "A String", # The name of the data disk where data for this range is stored. This name is local to the Google Cloud Platform project and uniquely identifies the disk within that project, for example "myproject-1014-104817-4c2-harness-0-disk-1".
"deliveryEndpoint": "A String", # The physical location of this range assignment to be used for streaming computation cross-worker message delivery.
"deprecatedPersistentDirectory": "A String", # DEPRECATED. The location of the persistent state for this range, as a persistent directory in the worker local filesystem.
"end": "A String", # The end (exclusive) of the key range.
"start": "A String", # The start (inclusive) of the key range.
},
],
"outputs": [ # The outputs from the computation.
{ # Describes a stream of data, either as input to be processed or as output of a streaming Dataflow job.
"customSourceLocation": { # Identifies the location of a custom souce. # The stream is a custom source.
"stateful": True or False, # Whether this source is stateful.
},
"pubsubLocation": { # Identifies a pubsub location to use for transferring data into or out of a streaming Dataflow job. # The stream is a pubsub stream.
"dropLateData": True or False, # Indicates whether the pipeline allows late-arriving data.
"dynamicDestinations": True or False, # If true, then this location represents dynamic topics.
"idLabel": "A String", # If set, contains a pubsub label from which to extract record ids. If left empty, record deduplication will be strictly best effort.
"subscription": "A String", # A pubsub subscription, in the form of "pubsub.googleapis.com/subscriptions//"
"timestampLabel": "A String", # If set, contains a pubsub label from which to extract record timestamps. If left empty, record timestamps will be generated upon arrival.
"topic": "A String", # A pubsub topic, in the form of "pubsub.googleapis.com/topics//"
"trackingSubscription": "A String", # If set, specifies the pubsub subscription that will be used for tracking custom time timestamps for watermark estimation.
"withAttributes": True or False, # If true, then the client has requested to get pubsub attributes.
},
"sideInputLocation": { # Identifies the location of a streaming side input. # The stream is a streaming side input.
"stateFamily": "A String", # Identifies the state family where this side input is stored.
"tag": "A String", # Identifies the particular side input within the streaming Dataflow job.
},
"streamingStageLocation": { # Identifies the location of a streaming computation stage, for stage-to-stage communication. # The stream is part of another computation within the current streaming Dataflow job.
"streamId": "A String", # Identifies the particular stream within the streaming Dataflow job.
},
},
],
"stateFamilies": [ # The state family values.
{ # State family configuration.
"isRead": True or False, # If true, this family corresponds to a read operation.
"stateFamily": "A String", # The state family value.
},
],
"systemStageName": "A String", # The system stage name.
},
],
"dataDiskAssignments": [ # The disks assigned to a streaming Dataflow job.
{ # Data disk assignment for a given VM instance.
"dataDisks": [ # Mounted data disks. The order is important a data disk's 0-based index in this list defines which persistent directory the disk is mounted to, for example the list of { "myproject-1014-104817-4c2-harness-0-disk-0" }, { "myproject-1014-104817-4c2-harness-0-disk-1" }.
"A String",
],
"vmInstance": "A String", # VM instance name the data disks mounted to, for example "myproject-1014-104817-4c2-harness-0".
},
],
"forwardingKeyBits": 42, # The size (in bits) of keys that will be assigned to source messages.
"persistentStateVersion": 42, # Version number for persistent state.
"userStageToComputationNameMap": { # Maps user stage names to stable computation names.
"a_key": "A String",
},
},
"workerHarnessPort": 42, # The TCP port used by the worker to communicate with the Dataflow worker harness.
},
},
],
}</pre>
</div>
<div class="method">
<code class="details" id="reportStatus">reportStatus(projectId, jobId, body=None, x__xgafv=None)</code>
<pre>Reports the status of dataflow WorkItems leased by a worker.
Args:
projectId: string, The project which owns the WorkItem's job. (required)
jobId: string, The job which the WorkItem is part of. (required)
body: object, The request body.
The object takes the form of:
{ # Request to report the status of WorkItems.
"currentWorkerTime": "A String", # The current timestamp at the worker.
"location": "A String", # The [regional endpoint] (https://cloud.google.com/dataflow/docs/concepts/regional-endpoints) that contains the WorkItem's job.
"unifiedWorkerRequest": { # Untranslated bag-of-bytes WorkProgressUpdateRequest from UnifiedWorker.
"a_key": "", # Properties of the object. Contains field @type with type URL.
},
"workItemStatuses": [ # The order is unimportant, except that the order of the WorkItemServiceState messages in the ReportWorkItemStatusResponse corresponds to the order of WorkItemStatus messages here.
{ # Conveys a worker's progress through the work described by a WorkItem.
"completed": True or False, # True if the WorkItem was completed (successfully or unsuccessfully).
"counterUpdates": [ # Worker output counters for this WorkItem.
{ # An update to a Counter sent from a worker.
"boolean": True or False, # Boolean value for And, Or.
"cumulative": True or False, # True if this counter is reported as the total cumulative aggregate value accumulated since the worker started working on this WorkItem. By default this is false, indicating that this counter is reported as a delta.
"distribution": { # A metric value representing a distribution. # Distribution data
"count": { # A representation of an int64, n, that is immune to precision loss when encoded in JSON. # The count of the number of elements present in the distribution.
"highBits": 42, # The high order bits, including the sign: n >> 32.
"lowBits": 42, # The low order bits: n & 0xffffffff.
},
"histogram": { # Histogram of value counts for a distribution. Buckets have an inclusive lower bound and exclusive upper bound and use "1,2,5 bucketing": The first bucket range is from [0,1) and all subsequent bucket boundaries are powers of ten multiplied by 1, 2, or 5. Thus, bucket boundaries are 0, 1, 2, 5, 10, 20, 50, 100, 200, 500, 1000, ... Negative values are not supported. # (Optional) Histogram of value counts for the distribution.
"bucketCounts": [ # Counts of values in each bucket. For efficiency, prefix and trailing buckets with count = 0 are elided. Buckets can store the full range of values of an unsigned long, with ULLONG_MAX falling into the 59th bucket with range [1e19, 2e19).
"A String",
],
"firstBucketOffset": 42, # Starting index of first stored bucket. The non-inclusive upper-bound of the ith bucket is given by: pow(10,(i-first_bucket_offset)/3) * (1,2,5)[(i-first_bucket_offset)%3]
},
"max": { # A representation of an int64, n, that is immune to precision loss when encoded in JSON. # The maximum value present in the distribution.
"highBits": 42, # The high order bits, including the sign: n >> 32.
"lowBits": 42, # The low order bits: n & 0xffffffff.
},
"min": { # A representation of an int64, n, that is immune to precision loss when encoded in JSON. # The minimum value present in the distribution.
"highBits": 42, # The high order bits, including the sign: n >> 32.
"lowBits": 42, # The low order bits: n & 0xffffffff.
},
"sum": { # A representation of an int64, n, that is immune to precision loss when encoded in JSON. # Use an int64 since we'd prefer the added precision. If overflow is a common problem we can detect it and use an additional int64 or a double.
"highBits": 42, # The high order bits, including the sign: n >> 32.
"lowBits": 42, # The low order bits: n & 0xffffffff.
},
"sumOfSquares": 3.14, # Use a double since the sum of squares is likely to overflow int64.
},
"floatingPoint": 3.14, # Floating point value for Sum, Max, Min.
"floatingPointList": { # A metric value representing a list of floating point numbers. # List of floating point numbers, for Set.
"elements": [ # Elements of the list.
3.14,
],
},
"floatingPointMean": { # A representation of a floating point mean metric contribution. # Floating point mean aggregation value for Mean.
"count": { # A representation of an int64, n, that is immune to precision loss when encoded in JSON. # The number of values being aggregated.
"highBits": 42, # The high order bits, including the sign: n >> 32.
"lowBits": 42, # The low order bits: n & 0xffffffff.
},
"sum": 3.14, # The sum of all values being aggregated.
},
"integer": { # A representation of an int64, n, that is immune to precision loss when encoded in JSON. # Integer value for Sum, Max, Min.
"highBits": 42, # The high order bits, including the sign: n >> 32.
"lowBits": 42, # The low order bits: n & 0xffffffff.
},
"integerGauge": { # A metric value representing temporal values of a variable. # Gauge data
"timestamp": "A String", # The time at which this value was measured. Measured as msecs from epoch.
"value": { # A representation of an int64, n, that is immune to precision loss when encoded in JSON. # The value of the variable represented by this gauge.
"highBits": 42, # The high order bits, including the sign: n >> 32.
"lowBits": 42, # The low order bits: n & 0xffffffff.
},
},
"integerList": { # A metric value representing a list of integers. # List of integers, for Set.
"elements": [ # Elements of the list.
{ # A representation of an int64, n, that is immune to precision loss when encoded in JSON.
"highBits": 42, # The high order bits, including the sign: n >> 32.
"lowBits": 42, # The low order bits: n & 0xffffffff.
},
],
},
"integerMean": { # A representation of an integer mean metric contribution. # Integer mean aggregation value for Mean.
"count": { # A representation of an int64, n, that is immune to precision loss when encoded in JSON. # The number of values being aggregated.
"highBits": 42, # The high order bits, including the sign: n >> 32.
"lowBits": 42, # The low order bits: n & 0xffffffff.
},
"sum": { # A representation of an int64, n, that is immune to precision loss when encoded in JSON. # The sum of all values being aggregated.
"highBits": 42, # The high order bits, including the sign: n >> 32.
"lowBits": 42, # The low order bits: n & 0xffffffff.
},
},
"internal": "", # Value for internally-defined counters used by the Dataflow service.
"nameAndKind": { # Basic metadata about a counter. # Counter name and aggregation type.
"kind": "A String", # Counter aggregation kind.
"name": "A String", # Name of the counter.
},
"shortId": "A String", # The service-generated short identifier for this counter. The short_id -> (name, metadata) mapping is constant for the lifetime of a job.
"stringList": { # A metric value representing a list of strings. # List of strings, for Set.
"elements": [ # Elements of the list.
"A String",
],
},
"structuredNameAndMetadata": { # A single message which encapsulates structured name and metadata for a given counter. # Counter structured name and metadata.
"metadata": { # CounterMetadata includes all static non-name non-value counter attributes. # Metadata associated with a counter
"description": "A String", # Human-readable description of the counter semantics.
"kind": "A String", # Counter aggregation kind.
"otherUnits": "A String", # A string referring to the unit type.
"standardUnits": "A String", # System defined Units, see above enum.
},
"name": { # Identifies a counter within a per-job namespace. Counters whose structured names are the same get merged into a single value for the job. # Structured name of the counter.
"componentStepName": "A String", # Name of the optimized step being executed by the workers.
"executionStepName": "A String", # Name of the stage. An execution step contains multiple component steps.
"inputIndex": 42, # Index of an input collection that's being read from/written to as a side input. The index identifies a step's side inputs starting by 1 (e.g. the first side input has input_index 1, the third has input_index 3). Side inputs are identified by a pair of (original_step_name, input_index). This field helps uniquely identify them.
"name": "A String", # Counter name. Not necessarily globally-unique, but unique within the context of the other fields. Required.
"origin": "A String", # One of the standard Origins defined above.
"originNamespace": "A String", # A string containing a more specific namespace of the counter's origin.
"originalRequestingStepName": "A String", # The step name requesting an operation, such as GBK. I.e. the ParDo causing a read/write from shuffle to occur, or a read from side inputs.
"originalStepName": "A String", # System generated name of the original step in the user's graph, before optimization.
"portion": "A String", # Portion of this counter, either key or value.
"workerId": "A String", # ID of a particular worker.
},
},
},
],
"dynamicSourceSplit": { # When a task splits using WorkItemStatus.dynamic_source_split, this message describes the two parts of the split relative to the description of the current task's input. # See documentation of stop_position.
"primary": { # Specification of one of the bundles produced as a result of splitting a Source (e.g. when executing a SourceSplitRequest, or when splitting an active task using WorkItemStatus.dynamic_source_split), relative to the source being split. # Primary part (continued to be processed by worker). Specified relative to the previously-current source. Becomes current.
"derivationMode": "A String", # What source to base the produced source on (if any).
"source": { # A source that records can be read and decoded from. # Specification of the source.
"baseSpecs": [ # While splitting, sources may specify the produced bundles as differences against another source, in order to save backend-side memory and allow bigger jobs. For details, see SourceSplitRequest. To support this use case, the full set of parameters of the source is logically obtained by taking the latest explicitly specified value of each parameter in the order: base_specs (later items win), spec (overrides anything in base_specs).
{
"a_key": "", # Properties of the object.
},
],
"codec": { # The codec to use to decode data read from the source.
"a_key": "", # Properties of the object.
},
"doesNotNeedSplitting": True or False, # Setting this value to true hints to the framework that the source doesn't need splitting, and using SourceSplitRequest on it would yield SOURCE_SPLIT_OUTCOME_USE_CURRENT. E.g. a file splitter may set this to true when splitting a single file into a set of byte ranges of appropriate size, and set this to false when splitting a filepattern into individual files. However, for efficiency, a file splitter may decide to produce file subranges directly from the filepattern to avoid a splitting round-trip. See SourceSplitRequest for an overview of the splitting process. This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"metadata": { # Metadata about a Source useful for automatically optimizing and tuning the pipeline, etc. # Optionally, metadata for this source can be supplied right away, avoiding a SourceGetMetadataOperation roundtrip (see SourceOperationRequest). This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"estimatedSizeBytes": "A String", # An estimate of the total size (in bytes) of the data that would be read from this source. This estimate is in terms of external storage size, before any decompression or other processing done by the reader.
"infinite": True or False, # Specifies that the size of this source is known to be infinite (this is a streaming source).
"producesSortedKeys": True or False, # Whether this source is known to produce key/value pairs with the (encoded) keys in lexicographically sorted order.
},
"spec": { # The source to read from, plus its parameters.
"a_key": "", # Properties of the object.
},
},
},
"residual": { # Specification of one of the bundles produced as a result of splitting a Source (e.g. when executing a SourceSplitRequest, or when splitting an active task using WorkItemStatus.dynamic_source_split), relative to the source being split. # Residual part (returned to the pool of work). Specified relative to the previously-current source.
"derivationMode": "A String", # What source to base the produced source on (if any).
"source": { # A source that records can be read and decoded from. # Specification of the source.
"baseSpecs": [ # While splitting, sources may specify the produced bundles as differences against another source, in order to save backend-side memory and allow bigger jobs. For details, see SourceSplitRequest. To support this use case, the full set of parameters of the source is logically obtained by taking the latest explicitly specified value of each parameter in the order: base_specs (later items win), spec (overrides anything in base_specs).
{
"a_key": "", # Properties of the object.
},
],
"codec": { # The codec to use to decode data read from the source.
"a_key": "", # Properties of the object.
},
"doesNotNeedSplitting": True or False, # Setting this value to true hints to the framework that the source doesn't need splitting, and using SourceSplitRequest on it would yield SOURCE_SPLIT_OUTCOME_USE_CURRENT. E.g. a file splitter may set this to true when splitting a single file into a set of byte ranges of appropriate size, and set this to false when splitting a filepattern into individual files. However, for efficiency, a file splitter may decide to produce file subranges directly from the filepattern to avoid a splitting round-trip. See SourceSplitRequest for an overview of the splitting process. This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"metadata": { # Metadata about a Source useful for automatically optimizing and tuning the pipeline, etc. # Optionally, metadata for this source can be supplied right away, avoiding a SourceGetMetadataOperation roundtrip (see SourceOperationRequest). This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"estimatedSizeBytes": "A String", # An estimate of the total size (in bytes) of the data that would be read from this source. This estimate is in terms of external storage size, before any decompression or other processing done by the reader.
"infinite": True or False, # Specifies that the size of this source is known to be infinite (this is a streaming source).
"producesSortedKeys": True or False, # Whether this source is known to produce key/value pairs with the (encoded) keys in lexicographically sorted order.
},
"spec": { # The source to read from, plus its parameters.
"a_key": "", # Properties of the object.
},
},
},
},
"errors": [ # Specifies errors which occurred during processing. If errors are provided, and completed = true, then the WorkItem is considered to have failed.
{ # The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors).
"code": 42, # The status code, which should be an enum value of google.rpc.Code.
"details": [ # A list of messages that carry the error details. There is a common set of message types for APIs to use.
{
"a_key": "", # Properties of the object. Contains field @type with type URL.
},
],
"message": "A String", # A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client.
},
],
"metricUpdates": [ # DEPRECATED in favor of counter_updates.
{ # Describes the state of a metric.
"cumulative": True or False, # True if this metric is reported as the total cumulative aggregate value accumulated since the worker started working on this WorkItem. By default this is false, indicating that this metric is reported as a delta that is not associated with any WorkItem.
"distribution": "", # A struct value describing properties of a distribution of numeric values.
"gauge": "", # A struct value describing properties of a Gauge. Metrics of gauge type show the value of a metric across time, and is aggregated based on the newest value.
"internal": "", # Worker-computed aggregate value for internal use by the Dataflow service.
"kind": "A String", # Metric aggregation kind. The possible metric aggregation kinds are "Sum", "Max", "Min", "Mean", "Set", "And", "Or", and "Distribution". The specified aggregation kind is case-insensitive. If omitted, this is not an aggregated value but instead a single metric sample value.
"meanCount": "", # Worker-computed aggregate value for the "Mean" aggregation kind. This holds the count of the aggregated values and is used in combination with mean_sum above to obtain the actual mean aggregate value. The only possible value type is Long.
"meanSum": "", # Worker-computed aggregate value for the "Mean" aggregation kind. This holds the sum of the aggregated values and is used in combination with mean_count below to obtain the actual mean aggregate value. The only possible value types are Long and Double.
"name": { # Identifies a metric, by describing the source which generated the metric. # Name of the metric.
"context": { # Zero or more labeled fields which identify the part of the job this metric is associated with, such as the name of a step or collection. For example, built-in counters associated with steps will have context['step'] = . Counters associated with PCollections in the SDK will have context['pcollection'] = .
"a_key": "A String",
},
"name": "A String", # Worker-defined metric name.
"origin": "A String", # Origin (namespace) of metric name. May be blank for user-define metrics; will be "dataflow" for metrics defined by the Dataflow service or SDK.
},
"scalar": "", # Worker-computed aggregate value for aggregation kinds "Sum", "Max", "Min", "And", and "Or". The possible value types are Long, Double, and Boolean.
"set": "", # Worker-computed aggregate value for the "Set" aggregation kind. The only possible value type is a list of Values whose type can be Long, Double, or String, according to the metric's type. All Values in the list must be of the same type.
"updateTime": "A String", # Timestamp associated with the metric value. Optional when workers are reporting work progress; it will be filled in responses from the metrics API.
},
],
"progress": { # Obsolete in favor of ApproximateReportedProgress and ApproximateSplitRequest. # DEPRECATED in favor of reported_progress.
"percentComplete": 3.14, # Obsolete.
"position": { # Position defines a position within a collection of data. The value can be either the end position, a key (used with ordered collections), a byte offset, or a record index. # Obsolete.
"byteOffset": "A String", # Position is a byte offset.
"concatPosition": { # A position that encapsulates an inner position and an index for the inner position. A ConcatPosition can be used by a reader of a source that encapsulates a set of other sources. # CloudPosition is a concat position.
"index": 42, # Index of the inner source.
"position": # Object with schema name: Position # Position within the inner source.
},
"end": True or False, # Position is past all other positions. Also useful for the end position of an unbounded range.
"key": "A String", # Position is a string key, ordered lexicographically.
"recordIndex": "A String", # Position is a record index.
"shufflePosition": "A String", # CloudPosition is a base64 encoded BatchShufflePosition (with FIXED sharding).
},
"remainingTime": "A String", # Obsolete.
},
"reportIndex": "A String", # The report index. When a WorkItem is leased, the lease will contain an initial report index. When a WorkItem's status is reported to the system, the report should be sent with that report index, and the response will contain the index the worker should use for the next report. Reports received with unexpected index values will be rejected by the service. In order to preserve idempotency, the worker should not alter the contents of a report, even if the worker must submit the same report multiple times before getting back a response. The worker should not submit a subsequent report until the response for the previous report had been received from the service.
"reportedProgress": { # A progress measurement of a WorkItem by a worker. # The worker's progress through this WorkItem.
"consumedParallelism": { # Represents the level of parallelism in a WorkItem's input, reported by the worker. # Total amount of parallelism in the portion of input of this task that has already been consumed and is no longer active. In the first two examples above (see remaining_parallelism), the value should be 29 or 2 respectively. The sum of remaining_parallelism and consumed_parallelism should equal the total amount of parallelism in this work item. If specified, must be finite.
"isInfinite": True or False, # Specifies whether the parallelism is infinite. If true, "value" is ignored. Infinite parallelism means the service will assume that the work item can always be split into more non-empty work items by dynamic splitting. This is a work-around for lack of support for infinity by the current JSON-based Java RPC stack.
"value": 3.14, # Specifies the level of parallelism in case it is finite.
},
"fractionConsumed": 3.14, # Completion as fraction of the input consumed, from 0.0 (beginning, nothing consumed), to 1.0 (end of the input, entire input consumed).
"position": { # Position defines a position within a collection of data. The value can be either the end position, a key (used with ordered collections), a byte offset, or a record index. # A Position within the work to represent a progress.
"byteOffset": "A String", # Position is a byte offset.
"concatPosition": { # A position that encapsulates an inner position and an index for the inner position. A ConcatPosition can be used by a reader of a source that encapsulates a set of other sources. # CloudPosition is a concat position.
"index": 42, # Index of the inner source.
"position": # Object with schema name: Position # Position within the inner source.
},
"end": True or False, # Position is past all other positions. Also useful for the end position of an unbounded range.
"key": "A String", # Position is a string key, ordered lexicographically.
"recordIndex": "A String", # Position is a record index.
"shufflePosition": "A String", # CloudPosition is a base64 encoded BatchShufflePosition (with FIXED sharding).
},
"remainingParallelism": { # Represents the level of parallelism in a WorkItem's input, reported by the worker. # Total amount of parallelism in the input of this task that remains, (i.e. can be delegated to this task and any new tasks via dynamic splitting). Always at least 1 for non-finished work items and 0 for finished. "Amount of parallelism" refers to how many non-empty parts of the input can be read in parallel. This does not necessarily equal number of records. An input that can be read in parallel down to the individual records is called "perfectly splittable". An example of non-perfectly parallelizable input is a block-compressed file format where a block of records has to be read as a whole, but different blocks can be read in parallel. Examples: * If we are processing record #30 (starting at 1) out of 50 in a perfectly splittable 50-record input, this value should be 21 (20 remaining + 1 current). * If we are reading through block 3 in a block-compressed file consisting of 5 blocks, this value should be 3 (since blocks 4 and 5 can be processed in parallel by new tasks via dynamic splitting and the current task remains processing block 3). * If we are reading through the last block in a block-compressed file, or reading or processing the last record in a perfectly splittable input, this value should be 1, because apart from the current task, no additional remainder can be split off.
"isInfinite": True or False, # Specifies whether the parallelism is infinite. If true, "value" is ignored. Infinite parallelism means the service will assume that the work item can always be split into more non-empty work items by dynamic splitting. This is a work-around for lack of support for infinity by the current JSON-based Java RPC stack.
"value": 3.14, # Specifies the level of parallelism in case it is finite.
},
},
"requestedLeaseDuration": "A String", # Amount of time the worker requests for its lease.
"sourceFork": { # DEPRECATED in favor of DynamicSourceSplit. # DEPRECATED in favor of dynamic_source_split.
"primary": { # DEPRECATED in favor of DerivedSource. # DEPRECATED
"derivationMode": "A String", # DEPRECATED
"source": { # A source that records can be read and decoded from. # DEPRECATED
"baseSpecs": [ # While splitting, sources may specify the produced bundles as differences against another source, in order to save backend-side memory and allow bigger jobs. For details, see SourceSplitRequest. To support this use case, the full set of parameters of the source is logically obtained by taking the latest explicitly specified value of each parameter in the order: base_specs (later items win), spec (overrides anything in base_specs).
{
"a_key": "", # Properties of the object.
},
],
"codec": { # The codec to use to decode data read from the source.
"a_key": "", # Properties of the object.
},
"doesNotNeedSplitting": True or False, # Setting this value to true hints to the framework that the source doesn't need splitting, and using SourceSplitRequest on it would yield SOURCE_SPLIT_OUTCOME_USE_CURRENT. E.g. a file splitter may set this to true when splitting a single file into a set of byte ranges of appropriate size, and set this to false when splitting a filepattern into individual files. However, for efficiency, a file splitter may decide to produce file subranges directly from the filepattern to avoid a splitting round-trip. See SourceSplitRequest for an overview of the splitting process. This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"metadata": { # Metadata about a Source useful for automatically optimizing and tuning the pipeline, etc. # Optionally, metadata for this source can be supplied right away, avoiding a SourceGetMetadataOperation roundtrip (see SourceOperationRequest). This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"estimatedSizeBytes": "A String", # An estimate of the total size (in bytes) of the data that would be read from this source. This estimate is in terms of external storage size, before any decompression or other processing done by the reader.
"infinite": True or False, # Specifies that the size of this source is known to be infinite (this is a streaming source).
"producesSortedKeys": True or False, # Whether this source is known to produce key/value pairs with the (encoded) keys in lexicographically sorted order.
},
"spec": { # The source to read from, plus its parameters.
"a_key": "", # Properties of the object.
},
},
},
"primarySource": { # Specification of one of the bundles produced as a result of splitting a Source (e.g. when executing a SourceSplitRequest, or when splitting an active task using WorkItemStatus.dynamic_source_split), relative to the source being split. # DEPRECATED
"derivationMode": "A String", # What source to base the produced source on (if any).
"source": { # A source that records can be read and decoded from. # Specification of the source.
"baseSpecs": [ # While splitting, sources may specify the produced bundles as differences against another source, in order to save backend-side memory and allow bigger jobs. For details, see SourceSplitRequest. To support this use case, the full set of parameters of the source is logically obtained by taking the latest explicitly specified value of each parameter in the order: base_specs (later items win), spec (overrides anything in base_specs).
{
"a_key": "", # Properties of the object.
},
],
"codec": { # The codec to use to decode data read from the source.
"a_key": "", # Properties of the object.
},
"doesNotNeedSplitting": True or False, # Setting this value to true hints to the framework that the source doesn't need splitting, and using SourceSplitRequest on it would yield SOURCE_SPLIT_OUTCOME_USE_CURRENT. E.g. a file splitter may set this to true when splitting a single file into a set of byte ranges of appropriate size, and set this to false when splitting a filepattern into individual files. However, for efficiency, a file splitter may decide to produce file subranges directly from the filepattern to avoid a splitting round-trip. See SourceSplitRequest for an overview of the splitting process. This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"metadata": { # Metadata about a Source useful for automatically optimizing and tuning the pipeline, etc. # Optionally, metadata for this source can be supplied right away, avoiding a SourceGetMetadataOperation roundtrip (see SourceOperationRequest). This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"estimatedSizeBytes": "A String", # An estimate of the total size (in bytes) of the data that would be read from this source. This estimate is in terms of external storage size, before any decompression or other processing done by the reader.
"infinite": True or False, # Specifies that the size of this source is known to be infinite (this is a streaming source).
"producesSortedKeys": True or False, # Whether this source is known to produce key/value pairs with the (encoded) keys in lexicographically sorted order.
},
"spec": { # The source to read from, plus its parameters.
"a_key": "", # Properties of the object.
},
},
},
"residual": { # DEPRECATED in favor of DerivedSource. # DEPRECATED
"derivationMode": "A String", # DEPRECATED
"source": { # A source that records can be read and decoded from. # DEPRECATED
"baseSpecs": [ # While splitting, sources may specify the produced bundles as differences against another source, in order to save backend-side memory and allow bigger jobs. For details, see SourceSplitRequest. To support this use case, the full set of parameters of the source is logically obtained by taking the latest explicitly specified value of each parameter in the order: base_specs (later items win), spec (overrides anything in base_specs).
{
"a_key": "", # Properties of the object.
},
],
"codec": { # The codec to use to decode data read from the source.
"a_key": "", # Properties of the object.
},
"doesNotNeedSplitting": True or False, # Setting this value to true hints to the framework that the source doesn't need splitting, and using SourceSplitRequest on it would yield SOURCE_SPLIT_OUTCOME_USE_CURRENT. E.g. a file splitter may set this to true when splitting a single file into a set of byte ranges of appropriate size, and set this to false when splitting a filepattern into individual files. However, for efficiency, a file splitter may decide to produce file subranges directly from the filepattern to avoid a splitting round-trip. See SourceSplitRequest for an overview of the splitting process. This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"metadata": { # Metadata about a Source useful for automatically optimizing and tuning the pipeline, etc. # Optionally, metadata for this source can be supplied right away, avoiding a SourceGetMetadataOperation roundtrip (see SourceOperationRequest). This field is meaningful only in the Source objects populated by the user (e.g. when filling in a DerivedSource). Source objects supplied by the framework to the user don't have this field populated.
"estimatedSizeBytes": "A String", # An estimate of the total size (in bytes) of the data that would be read from this source. This estimate is in terms of external storage size, before any decompression or other processing done by the reader.
"infinite": True or False, # Specifies that the size of this source is known to be infinite (this is a streaming source).
"producesSortedKeys": True or False, # Whether this source is known to produce key/value pairs with the (encoded) keys in lexicographically sorted order.
},
"spec": { # The source to read from, plus its parameters.
"a_key": "", # Properties of the object.
},
},
},
"residualSource": { # Specification of one of the bundles produced as a result of splitting a Source (e.g. when executing a SourceSplitRequest, or when splitting an active task using WorkItemStatus.dynamic_source_split), relative to the source being split. # DEPRECATED
"derivationMode": "A String", # What source to base the produced source on (if any).