forked from datahub-project/datahub
/
sql_common.py
1237 lines (1099 loc) · 44.2 KB
/
sql_common.py
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
import datetime
import logging
import traceback
from collections import OrderedDict
from dataclasses import dataclass, field
from typing import (
TYPE_CHECKING,
Any,
Callable,
Dict,
Iterable,
List,
Optional,
Set,
Tuple,
Type,
Union,
)
import sqlalchemy.dialects.postgresql.base
from sqlalchemy import create_engine, inspect
from sqlalchemy.engine.reflection import Inspector
from sqlalchemy.exc import ProgrammingError
from sqlalchemy.sql import sqltypes as types
from sqlalchemy.types import TypeDecorator, TypeEngine
from datahub.emitter.mce_builder import (
make_data_platform_urn,
make_dataplatform_instance_urn,
make_dataset_urn_with_platform_instance,
make_tag_urn,
)
from datahub.emitter.mcp import MetadataChangeProposalWrapper
from datahub.ingestion.api.common import PipelineContext
from datahub.ingestion.api.workunit import MetadataWorkUnit
from datahub.ingestion.source.common.subtypes import (
DatasetContainerSubTypes,
DatasetSubTypes,
)
from datahub.ingestion.source.sql.sql_config import SQLAlchemyConfig
from datahub.ingestion.source.sql.sql_utils import (
add_table_to_schema_container,
gen_database_container,
gen_database_key,
gen_schema_container,
gen_schema_key,
get_domain_wu,
)
from datahub.ingestion.source.state.sql_common_state import (
BaseSQLAlchemyCheckpointState,
)
from datahub.ingestion.source.state.stale_entity_removal_handler import (
StaleEntityRemovalHandler,
StaleEntityRemovalSourceReport,
)
from datahub.ingestion.source.state.stateful_ingestion_base import (
StatefulIngestionSourceBase,
)
from datahub.metadata.com.linkedin.pegasus2avro.common import StatusClass
from datahub.metadata.com.linkedin.pegasus2avro.dataset import UpstreamLineage
from datahub.metadata.com.linkedin.pegasus2avro.metadata.snapshot import DatasetSnapshot
from datahub.metadata.com.linkedin.pegasus2avro.mxe import MetadataChangeEvent
from datahub.metadata.com.linkedin.pegasus2avro.schema import (
ArrayTypeClass,
BooleanTypeClass,
BytesTypeClass,
DateTypeClass,
EnumTypeClass,
ForeignKeyConstraint,
MySqlDDL,
NullTypeClass,
NumberTypeClass,
RecordTypeClass,
SchemaField,
SchemaFieldDataType,
SchemaMetadata,
StringTypeClass,
TimeTypeClass,
)
from datahub.metadata.schema_classes import (
ChangeTypeClass,
DataPlatformInstanceClass,
DatasetLineageTypeClass,
DatasetPropertiesClass,
GlobalTagsClass,
SubTypesClass,
TagAssociationClass,
UpstreamClass,
ViewPropertiesClass,
)
from datahub.telemetry import telemetry
from datahub.utilities.lossy_collections import LossyList
from datahub.utilities.registries.domain_registry import DomainRegistry
from datahub.utilities.source_helpers import (
auto_stale_entity_removal,
auto_status_aspect,
)
from datahub.utilities.sqlalchemy_query_combiner import SQLAlchemyQueryCombinerReport
if TYPE_CHECKING:
from datahub.ingestion.source.ge_data_profiler import (
DatahubGEProfiler,
GEProfilerRequest,
)
logger: logging.Logger = logging.getLogger(__name__)
MISSING_COLUMN_INFO = "missing column information"
def _platform_alchemy_uri_tester_gen(
platform: str, opt_starts_with: Optional[str] = None
) -> Tuple[str, Callable[[str], bool]]:
return platform, lambda x: x.startswith(
platform if not opt_starts_with else opt_starts_with
)
PLATFORM_TO_SQLALCHEMY_URI_TESTER_MAP: Dict[str, Callable[[str], bool]] = OrderedDict(
[
_platform_alchemy_uri_tester_gen("athena", "awsathena"),
_platform_alchemy_uri_tester_gen("bigquery"),
_platform_alchemy_uri_tester_gen("clickhouse"),
_platform_alchemy_uri_tester_gen("druid"),
_platform_alchemy_uri_tester_gen("hana"),
_platform_alchemy_uri_tester_gen("hive"),
_platform_alchemy_uri_tester_gen("mongodb"),
_platform_alchemy_uri_tester_gen("mssql"),
_platform_alchemy_uri_tester_gen("mysql"),
_platform_alchemy_uri_tester_gen("oracle"),
_platform_alchemy_uri_tester_gen("pinot"),
_platform_alchemy_uri_tester_gen("presto"),
(
"redshift",
lambda x: (
x.startswith(("jdbc:postgres:", "postgresql"))
and x.find("redshift.amazonaws") > 0
)
or x.startswith("redshift"),
),
# Don't move this before redshift.
_platform_alchemy_uri_tester_gen("postgres", "postgresql"),
_platform_alchemy_uri_tester_gen("snowflake"),
_platform_alchemy_uri_tester_gen("trino"),
_platform_alchemy_uri_tester_gen("vertica"),
]
)
def get_platform_from_sqlalchemy_uri(sqlalchemy_uri: str) -> str:
for platform, tester in PLATFORM_TO_SQLALCHEMY_URI_TESTER_MAP.items():
if tester(sqlalchemy_uri):
return platform
return "external"
@dataclass
class SQLSourceReport(StaleEntityRemovalSourceReport):
tables_scanned: int = 0
views_scanned: int = 0
entities_profiled: int = 0
filtered: LossyList[str] = field(default_factory=LossyList)
query_combiner: Optional[SQLAlchemyQueryCombinerReport] = None
def report_entity_scanned(self, name: str, ent_type: str = "table") -> None:
"""
Entity could be a view or a table
"""
if ent_type == "table":
self.tables_scanned += 1
elif ent_type == "view":
self.views_scanned += 1
else:
raise KeyError(f"Unknown entity {ent_type}.")
def report_entity_profiled(self, name: str) -> None:
self.entities_profiled += 1
def report_dropped(self, ent_name: str) -> None:
self.filtered.append(ent_name)
def report_from_query_combiner(
self, query_combiner_report: SQLAlchemyQueryCombinerReport
) -> None:
self.query_combiner = query_combiner_report
class SqlWorkUnit(MetadataWorkUnit):
pass
_field_type_mapping: Dict[Type[TypeEngine], Type] = {
types.Integer: NumberTypeClass,
types.Numeric: NumberTypeClass,
types.Boolean: BooleanTypeClass,
types.Enum: EnumTypeClass,
types._Binary: BytesTypeClass,
types.LargeBinary: BytesTypeClass,
types.PickleType: BytesTypeClass,
types.ARRAY: ArrayTypeClass,
types.String: StringTypeClass,
types.Date: DateTypeClass,
types.DATE: DateTypeClass,
types.Time: TimeTypeClass,
types.DateTime: TimeTypeClass,
types.DATETIME: TimeTypeClass,
types.TIMESTAMP: TimeTypeClass,
types.JSON: RecordTypeClass,
# Because the postgresql dialect is used internally by many other dialects,
# we add some postgres types here. This is ok to do because the postgresql
# dialect is built-in to sqlalchemy.
sqlalchemy.dialects.postgresql.base.BYTEA: BytesTypeClass,
sqlalchemy.dialects.postgresql.base.DOUBLE_PRECISION: NumberTypeClass,
sqlalchemy.dialects.postgresql.base.INET: StringTypeClass,
sqlalchemy.dialects.postgresql.base.MACADDR: StringTypeClass,
sqlalchemy.dialects.postgresql.base.MONEY: NumberTypeClass,
sqlalchemy.dialects.postgresql.base.OID: StringTypeClass,
sqlalchemy.dialects.postgresql.base.REGCLASS: BytesTypeClass,
sqlalchemy.dialects.postgresql.base.TIMESTAMP: TimeTypeClass,
sqlalchemy.dialects.postgresql.base.TIME: TimeTypeClass,
sqlalchemy.dialects.postgresql.base.INTERVAL: TimeTypeClass,
sqlalchemy.dialects.postgresql.base.BIT: BytesTypeClass,
sqlalchemy.dialects.postgresql.base.UUID: StringTypeClass,
sqlalchemy.dialects.postgresql.base.TSVECTOR: BytesTypeClass,
sqlalchemy.dialects.postgresql.base.ENUM: EnumTypeClass,
# When SQLAlchemy is unable to map a type into its internal hierarchy, it
# assigns the NullType by default. We want to carry this warning through.
types.NullType: NullTypeClass,
}
_known_unknown_field_types: Set[Type[TypeEngine]] = {
types.Interval,
types.CLOB,
}
def register_custom_type(tp: Type[TypeEngine], output: Optional[Type] = None) -> None:
if output:
_field_type_mapping[tp] = output
else:
_known_unknown_field_types.add(tp)
class _CustomSQLAlchemyDummyType(TypeDecorator):
impl = types.LargeBinary
def make_sqlalchemy_type(name: str) -> Type[TypeEngine]:
# This usage of type() dynamically constructs a class.
# See https://stackoverflow.com/a/15247202/5004662 and
# https://docs.python.org/3/library/functions.html#type.
sqlalchemy_type: Type[TypeEngine] = type(
name,
(_CustomSQLAlchemyDummyType,),
{
"__repr__": lambda self: f"{name}()",
},
)
return sqlalchemy_type
def get_column_type(
sql_report: SQLSourceReport, dataset_name: str, column_type: Any
) -> SchemaFieldDataType:
"""
Maps SQLAlchemy types (https://docs.sqlalchemy.org/en/13/core/type_basics.html) to corresponding schema types
"""
TypeClass: Optional[Type] = None
for sql_type in _field_type_mapping.keys():
if isinstance(column_type, sql_type):
TypeClass = _field_type_mapping[sql_type]
break
if TypeClass is None:
for sql_type in _known_unknown_field_types:
if isinstance(column_type, sql_type):
TypeClass = NullTypeClass
break
if TypeClass is None:
sql_report.report_warning(
dataset_name, f"unable to map type {column_type!r} to metadata schema"
)
TypeClass = NullTypeClass
return SchemaFieldDataType(type=TypeClass())
def get_schema_metadata(
sql_report: SQLSourceReport,
dataset_name: str,
platform: str,
columns: List[dict],
pk_constraints: Optional[dict] = None,
foreign_keys: Optional[List[ForeignKeyConstraint]] = None,
canonical_schema: List[SchemaField] = [],
) -> SchemaMetadata:
schema_metadata = SchemaMetadata(
schemaName=dataset_name,
platform=make_data_platform_urn(platform),
version=0,
hash="",
platformSchema=MySqlDDL(tableSchema=""),
fields=canonical_schema,
)
if foreign_keys is not None and foreign_keys != []:
schema_metadata.foreignKeys = foreign_keys
return schema_metadata
# config flags to emit telemetry for
config_options_to_report = [
"include_views",
"include_tables",
]
class SQLAlchemySource(StatefulIngestionSourceBase):
"""A Base class for all SQL Sources that use SQLAlchemy to extend"""
def __init__(self, config: SQLAlchemyConfig, ctx: PipelineContext, platform: str):
super(SQLAlchemySource, self).__init__(config, ctx)
self.config = config
self.platform = platform
self.report: SQLSourceReport = SQLSourceReport()
# Create and register the stateful ingestion use-case handlers.
self.stale_entity_removal_handler = StaleEntityRemovalHandler(
source=self,
config=self.config,
state_type_class=BaseSQLAlchemyCheckpointState,
pipeline_name=self.ctx.pipeline_name,
run_id=self.ctx.run_id,
)
config_report = {
config_option: config.dict().get(config_option)
for config_option in config_options_to_report
}
config_report = {
**config_report,
"profiling_enabled": config.profiling.enabled,
"platform": platform,
}
telemetry.telemetry_instance.ping(
"sql_config",
config_report,
)
if config.profiling.enabled:
telemetry.telemetry_instance.ping(
"sql_profiling_config",
config.profiling.config_for_telemetry(),
)
self.domain_registry: Optional[DomainRegistry] = None
if self.config.domain:
self.domain_registry = DomainRegistry(
cached_domains=[k for k in self.config.domain], graph=self.ctx.graph
)
def warn(self, log: logging.Logger, key: str, reason: str) -> None:
self.report.report_warning(key, reason)
log.warning(f"{key} => {reason}")
def error(self, log: logging.Logger, key: str, reason: str) -> None:
self.report.report_failure(key, reason)
log.error(f"{key} => {reason}")
def get_inspectors(self) -> Iterable[Inspector]:
# This method can be overridden in the case that you want to dynamically
# run on multiple databases.
url = self.config.get_sql_alchemy_url()
logger.debug(f"sql_alchemy_url={url}")
engine = create_engine(url, **self.config.options)
with engine.connect() as conn:
inspector = inspect(conn)
yield inspector
def get_db_name(self, inspector: Inspector) -> str:
engine = inspector.engine
if engine and hasattr(engine, "url") and hasattr(engine.url, "database"):
return str(engine.url.database).strip('"').lower()
else:
raise Exception("Unable to get database name from Sqlalchemy inspector")
def get_schema_names(self, inspector):
return inspector.get_schema_names()
def get_allowed_schemas(self, inspector: Inspector, db_name: str) -> Iterable[str]:
# this function returns the schema names which are filtered by schema_pattern.
for schema in self.get_schema_names(inspector):
if not self.config.schema_pattern.allowed(schema):
self.report.report_dropped(f"{schema}.*")
continue
else:
self.add_information_for_schema(inspector, schema)
yield schema
def gen_database_containers(
self,
database: str,
extra_properties: Optional[Dict[str, Any]] = None,
) -> Iterable[MetadataWorkUnit]:
database_container_key = gen_database_key(
database,
platform=self.platform,
platform_instance=self.config.platform_instance,
env=self.config.env,
)
yield from gen_database_container(
database=database,
database_container_key=database_container_key,
sub_types=[DatasetContainerSubTypes.DATABASE],
domain_registry=self.domain_registry,
domain_config=self.config.domain,
report=self.report,
extra_properties=extra_properties,
)
def gen_schema_containers(
self,
schema: str,
database: str,
extra_properties: Optional[Dict[str, Any]] = None,
) -> Iterable[MetadataWorkUnit]:
database_container_key = gen_database_key(
database,
platform=self.platform,
platform_instance=self.config.platform_instance,
env=self.config.env,
)
schema_container_key = gen_schema_key(
db_name=database,
schema=schema,
platform=self.platform,
platform_instance=self.config.platform_instance,
env=self.config.env,
)
yield from gen_schema_container(
database=database,
schema=schema,
schema_container_key=schema_container_key,
database_container_key=database_container_key,
sub_types=[DatasetContainerSubTypes.SCHEMA],
domain_registry=self.domain_registry,
domain_config=self.config.domain,
report=self.report,
extra_properties=extra_properties,
)
def add_table_to_schema_container(
self,
dataset_urn: str,
db_name: str,
schema: str,
) -> Iterable[MetadataWorkUnit]:
schema_container_key = gen_schema_key(
db_name=db_name,
schema=schema,
platform=self.platform,
platform_instance=self.config.platform_instance,
env=self.config.env,
)
yield from add_table_to_schema_container(
dataset_urn=dataset_urn,
parent_container_key=schema_container_key,
report=self.report,
)
def get_workunits_internal(self) -> Iterable[Union[MetadataWorkUnit, SqlWorkUnit]]:
sql_config = self.config
if logger.isEnabledFor(logging.DEBUG):
# If debug logging is enabled, we also want to echo each SQL query issued.
sql_config.options.setdefault("echo", True)
# Extra default SQLAlchemy option for better connection pooling and threading.
# https://docs.sqlalchemy.org/en/14/core/pooling.html#sqlalchemy.pool.QueuePool.params.max_overflow
if sql_config.profiling.enabled:
sql_config.options.setdefault(
"max_overflow", sql_config.profiling.max_workers
)
for inspector in self.get_inspectors():
profiler = None
profile_requests: List["GEProfilerRequest"] = []
if sql_config.profiling.enabled:
profiler = self.get_profiler_instance(inspector)
db_name = self.get_db_name(inspector)
yield from self.gen_database_containers(
database=db_name,
)
for schema in self.get_allowed_schemas(inspector, db_name):
self.add_information_for_schema(inspector, schema)
yield from self.gen_schema_containers(
database=db_name,
schema=schema,
extra_properties=self.get_schema_properties(
inspector=inspector, schema=schema, database=db_name
),
)
if sql_config.include_tables:
yield from self.loop_tables(inspector, schema, sql_config)
if sql_config.include_views:
yield from self.loop_views(inspector, schema, sql_config)
if profiler:
profile_requests += list(
self.loop_profiler_requests(inspector, schema, sql_config)
)
if profiler and profile_requests:
yield from self.loop_profiler(
profile_requests, profiler, platform=self.platform
)
def get_workunits(self) -> Iterable[MetadataWorkUnit]:
return auto_stale_entity_removal(
self.stale_entity_removal_handler,
auto_status_aspect(self.get_workunits_internal()),
)
def standardize_schema_table_names(
self, schema: str, entity: str
) -> Tuple[str, str]:
# Some SQLAlchemy dialects need a standardization step to clean the schema
# and table names. See BigQuery for an example of when this is useful.
return schema, entity
def get_identifier(
self, *, schema: str, entity: str, inspector: Inspector, **kwargs: Any
) -> str:
# Many SQLAlchemy dialects have three-level hierarchies. This method, which
# subclasses can override, enables them to modify the identifiers as needed.
if hasattr(self.config, "get_identifier"):
# This path is deprecated and will eventually be removed.
return self.config.get_identifier(schema=schema, table=entity) # type: ignore
else:
return f"{schema}.{entity}"
def get_foreign_key_metadata(
self,
dataset_urn: str,
schema: str,
fk_dict: Dict[str, str],
inspector: Inspector,
) -> ForeignKeyConstraint:
referred_schema: Optional[str] = fk_dict.get("referred_schema")
if not referred_schema:
referred_schema = schema
referred_dataset_name = self.get_identifier(
schema=referred_schema,
entity=fk_dict["referred_table"],
inspector=inspector,
)
source_fields = [
f"urn:li:schemaField:({dataset_urn},{f})"
for f in fk_dict["constrained_columns"]
]
foreign_dataset = make_dataset_urn_with_platform_instance(
platform=self.platform,
name=referred_dataset_name,
platform_instance=self.config.platform_instance,
env=self.config.env,
)
foreign_fields = [
f"urn:li:schemaField:({foreign_dataset},{f})"
for f in fk_dict["referred_columns"]
]
return ForeignKeyConstraint(
fk_dict["name"], foreign_fields, source_fields, foreign_dataset
)
def normalise_dataset_name(self, dataset_name: str) -> str:
return dataset_name
def loop_tables( # noqa: C901
self,
inspector: Inspector,
schema: str,
sql_config: SQLAlchemyConfig,
) -> Iterable[Union[SqlWorkUnit, MetadataWorkUnit]]:
tables_seen: Set[str] = set()
try:
for table in inspector.get_table_names(schema):
schema, table = self.standardize_schema_table_names(
schema=schema, entity=table
)
dataset_name = self.get_identifier(
schema=schema, entity=table, inspector=inspector
)
dataset_name = self.normalise_dataset_name(dataset_name)
if dataset_name not in tables_seen:
tables_seen.add(dataset_name)
else:
logger.debug(f"{dataset_name} has already been seen, skipping...")
continue
self.report.report_entity_scanned(dataset_name, ent_type="table")
if not sql_config.table_pattern.allowed(dataset_name):
self.report.report_dropped(dataset_name)
continue
try:
yield from self._process_table(
dataset_name, inspector, schema, table, sql_config
)
except Exception as e:
logger.warning(
f"Unable to ingest {schema}.{table} due to an exception.\n {traceback.format_exc()}"
)
self.report.report_warning(
f"{schema}.{table}", f"Ingestion error: {e}"
)
except Exception as e:
self.report.report_failure(f"{schema}", f"Tables error: {e}")
def add_information_for_schema(self, inspector: Inspector, schema: str) -> None:
pass
def get_extra_tags(
self, inspector: Inspector, schema: str, table: str
) -> Optional[Dict[str, List[str]]]:
return None
def _process_table(
self,
dataset_name: str,
inspector: Inspector,
schema: str,
table: str,
sql_config: SQLAlchemyConfig,
) -> Iterable[Union[SqlWorkUnit, MetadataWorkUnit]]:
columns = self._get_columns(dataset_name, inspector, schema, table)
dataset_urn = make_dataset_urn_with_platform_instance(
self.platform,
dataset_name,
self.config.platform_instance,
self.config.env,
)
dataset_snapshot = DatasetSnapshot(
urn=dataset_urn,
aspects=[StatusClass(removed=False)],
)
description, properties, location_urn = self.get_table_properties(
inspector, schema, table
)
# Tablename might be different from the real table if we ran some normalisation ont it.
# Getting normalized table name from the dataset_name
# Table is the last item in the dataset name
normalised_table = table
splits = dataset_name.split(".")
if splits:
normalised_table = splits[-1]
if properties and normalised_table != table:
properties["original_table_name"] = table
dataset_properties = DatasetPropertiesClass(
name=normalised_table,
description=description,
customProperties=properties,
)
dataset_snapshot.aspects.append(dataset_properties)
if self.config.include_table_location_lineage and location_urn:
external_upstream_table = UpstreamClass(
dataset=location_urn,
type=DatasetLineageTypeClass.COPY,
)
lineage_mcpw = MetadataChangeProposalWrapper(
entityType="dataset",
changeType=ChangeTypeClass.UPSERT,
entityUrn=dataset_snapshot.urn,
aspectName="upstreamLineage",
aspect=UpstreamLineage(upstreams=[external_upstream_table]),
)
lineage_wu = MetadataWorkUnit(
id=f"{self.platform}-{lineage_mcpw.entityUrn}-{lineage_mcpw.aspectName}",
mcp=lineage_mcpw,
)
self.report.report_workunit(lineage_wu)
yield lineage_wu
extra_tags = self.get_extra_tags(inspector, schema, table)
pk_constraints: dict = inspector.get_pk_constraint(table, schema)
foreign_keys = self._get_foreign_keys(dataset_urn, inspector, schema, table)
schema_fields = self.get_schema_fields(
dataset_name, columns, pk_constraints, tags=extra_tags
)
schema_metadata = get_schema_metadata(
self.report,
dataset_name,
self.platform,
columns,
pk_constraints,
foreign_keys,
schema_fields,
)
dataset_snapshot.aspects.append(schema_metadata)
db_name = self.get_db_name(inspector)
yield from self.add_table_to_schema_container(
dataset_urn=dataset_urn, db_name=db_name, schema=schema
)
mce = MetadataChangeEvent(proposedSnapshot=dataset_snapshot)
wu = SqlWorkUnit(id=dataset_name, mce=mce)
self.report.report_workunit(wu)
yield wu
dpi_aspect = self.get_dataplatform_instance_aspect(dataset_urn=dataset_urn)
if dpi_aspect:
yield dpi_aspect
subtypes_aspect = MetadataWorkUnit(
id=f"{dataset_name}-subtypes",
mcp=MetadataChangeProposalWrapper(
entityType="dataset",
changeType=ChangeTypeClass.UPSERT,
entityUrn=dataset_urn,
aspectName="subTypes",
aspect=SubTypesClass(typeNames=[DatasetSubTypes.TABLE]),
),
)
self.report.report_workunit(subtypes_aspect)
yield subtypes_aspect
if self.config.domain:
assert self.domain_registry
yield from get_domain_wu(
dataset_name=dataset_name,
entity_urn=dataset_urn,
domain_config=sql_config.domain,
domain_registry=self.domain_registry,
report=self.report,
)
def get_database_properties(
self, inspector: Inspector, database: str
) -> Optional[Dict[str, str]]:
return None
def get_schema_properties(
self, inspector: Inspector, database: str, schema: str
) -> Optional[Dict[str, str]]:
return None
def get_table_properties(
self, inspector: Inspector, schema: str, table: str
) -> Tuple[Optional[str], Dict[str, str], Optional[str]]:
description: Optional[str] = None
properties: Dict[str, str] = {}
# The location cannot be fetched generically, but subclasses may override
# this method and provide a location.
location: Optional[str] = None
try:
# SQLAlchemy stubs are incomplete and missing this method.
# PR: https://github.com/dropbox/sqlalchemy-stubs/pull/223.
table_info: dict = inspector.get_table_comment(table, schema) # type: ignore
except NotImplementedError:
return description, properties, location
except ProgrammingError as pe:
# Snowflake needs schema names quoted when fetching table comments.
logger.debug(
f"Encountered ProgrammingError. Retrying with quoted schema name for schema {schema} and table {table}",
pe,
)
table_info: dict = inspector.get_table_comment(table, f'"{schema}"') # type: ignore
description = table_info.get("text")
if type(description) is tuple:
# Handling for value type tuple which is coming for dialect 'db2+ibm_db'
description = table_info["text"][0]
# The "properties" field is a non-standard addition to SQLAlchemy's interface.
properties = table_info.get("properties", {})
return description, properties, location
def get_dataplatform_instance_aspect(
self, dataset_urn: str
) -> Optional[SqlWorkUnit]:
# If we are a platform instance based source, emit the instance aspect
if self.config.platform_instance:
mcp = MetadataChangeProposalWrapper(
entityType="dataset",
changeType=ChangeTypeClass.UPSERT,
entityUrn=dataset_urn,
aspectName="dataPlatformInstance",
aspect=DataPlatformInstanceClass(
platform=make_data_platform_urn(self.platform),
instance=make_dataplatform_instance_urn(
self.platform, self.config.platform_instance
),
),
)
wu = SqlWorkUnit(id=f"{dataset_urn}-dataPlatformInstance", mcp=mcp)
self.report.report_workunit(wu)
return wu
else:
return None
def _get_columns(
self, dataset_name: str, inspector: Inspector, schema: str, table: str
) -> List[dict]:
columns = []
try:
columns = inspector.get_columns(table, schema)
if len(columns) == 0:
self.report.report_warning(MISSING_COLUMN_INFO, dataset_name)
except Exception as e:
self.report.report_warning(
dataset_name,
f"unable to get column information due to an error -> {e}",
)
return columns
def _get_foreign_keys(
self, dataset_urn: str, inspector: Inspector, schema: str, table: str
) -> List[ForeignKeyConstraint]:
try:
foreign_keys = [
self.get_foreign_key_metadata(dataset_urn, schema, fk_rec, inspector)
for fk_rec in inspector.get_foreign_keys(table, schema)
]
except KeyError:
# certain databases like MySQL cause issues due to lower-case/upper-case irregularities
logger.debug(
f"{dataset_urn}: failure in foreign key extraction... skipping"
)
foreign_keys = []
return foreign_keys
def get_schema_fields(
self,
dataset_name: str,
columns: List[dict],
pk_constraints: Optional[dict] = None,
tags: Optional[Dict[str, List[str]]] = None,
) -> List[SchemaField]:
canonical_schema = []
for column in columns:
column_tags: Optional[List[str]] = None
if tags:
column_tags = tags.get(column["name"], [])
fields = self.get_schema_fields_for_column(
dataset_name, column, pk_constraints, tags=column_tags
)
canonical_schema.extend(fields)
return canonical_schema
def get_schema_fields_for_column(
self,
dataset_name: str,
column: dict,
pk_constraints: Optional[dict] = None,
tags: Optional[List[str]] = None,
) -> List[SchemaField]:
gtc: Optional[GlobalTagsClass] = None
if tags:
tags_str = [make_tag_urn(t) for t in tags]
tags_tac = [TagAssociationClass(t) for t in tags_str]
gtc = GlobalTagsClass(tags_tac)
field = SchemaField(
fieldPath=column["name"],
type=get_column_type(self.report, dataset_name, column["type"]),
nativeDataType=column.get("full_type", repr(column["type"])),
description=column.get("comment", None),
nullable=column["nullable"],
recursive=False,
globalTags=gtc,
)
if (
pk_constraints is not None
and isinstance(pk_constraints, dict) # some dialects (hive) return list
and column["name"] in pk_constraints.get("constrained_columns", [])
):
field.isPartOfKey = True
return [field]
def loop_views(
self,
inspector: Inspector,
schema: str,
sql_config: SQLAlchemyConfig,
) -> Iterable[Union[SqlWorkUnit, MetadataWorkUnit]]:
try:
for view in inspector.get_view_names(schema):
schema, view = self.standardize_schema_table_names(
schema=schema, entity=view
)
dataset_name = self.get_identifier(
schema=schema, entity=view, inspector=inspector
)
dataset_name = self.normalise_dataset_name(dataset_name)
self.report.report_entity_scanned(dataset_name, ent_type="view")
if not sql_config.view_pattern.allowed(dataset_name):
self.report.report_dropped(dataset_name)
continue
try:
yield from self._process_view(
dataset_name=dataset_name,
inspector=inspector,
schema=schema,
view=view,
sql_config=sql_config,
)
except Exception as e:
logger.warning(
f"Unable to ingest view {schema}.{view} due to an exception.\n {traceback.format_exc()}"
)
self.report.report_warning(
f"{schema}.{view}", f"Ingestion error: {e}"
)
except Exception as e:
self.report.report_failure(f"{schema}", f"Views error: {e}")
def _process_view(
self,
dataset_name: str,
inspector: Inspector,
schema: str,
view: str,
sql_config: SQLAlchemyConfig,
) -> Iterable[Union[SqlWorkUnit, MetadataWorkUnit]]:
try:
columns = inspector.get_columns(view, schema)
except KeyError:
# For certain types of views, we are unable to fetch the list of columns.
self.report.report_warning(
dataset_name, "unable to get schema for this view"
)
schema_metadata = None
else:
schema_fields = self.get_schema_fields(dataset_name, columns)
schema_metadata = get_schema_metadata(
self.report,
dataset_name,
self.platform,
columns,
canonical_schema=schema_fields,
)
description, properties, _ = self.get_table_properties(inspector, schema, view)
try:
view_definition = inspector.get_view_definition(view, schema)
if view_definition is None:
view_definition = ""
else:
# Some dialects return a TextClause instead of a raw string,
# so we need to convert them to a string.
view_definition = str(view_definition)
except NotImplementedError:
view_definition = ""
properties["view_definition"] = view_definition
properties["is_view"] = "True"
dataset_urn = make_dataset_urn_with_platform_instance(
self.platform,
dataset_name,
self.config.platform_instance,
self.config.env,
)
dataset_snapshot = DatasetSnapshot(
urn=dataset_urn,
aspects=[StatusClass(removed=False)],
)
db_name = self.get_db_name(inspector)
yield from self.add_table_to_schema_container(
dataset_urn=dataset_urn,
db_name=db_name,
schema=schema,
)
dataset_properties = DatasetPropertiesClass(
name=view,
description=description,