acryl-datahub 0.15.0.1rc16__py3-none-any.whl → 0.15.0.2__py3-none-any.whl

datahub/ingestion/source/snowflake/snowflake_v2.py

@@ -23,7 +23,6 @@ from datahub.ingestion.api.incremental_properties_helper import (
 from datahub.ingestion.api.source import (
     CapabilityReport,
     MetadataWorkUnitProcessor,
-    Source,
     SourceCapability,
     SourceReport,
     TestableSource,
@@ -212,9 +211,9 @@ class SnowflakeV2Source(
 
         self.usage_extractor: Optional[SnowflakeUsageExtractor] = None
         if self.config.include_usage_stats or self.config.include_operational_stats:
-            redundant_usage_run_skip_handler: Optional[
-                RedundantUsageRunSkipHandler
-            ] = None
+            redundant_usage_run_skip_handler: Optional[RedundantUsageRunSkipHandler] = (
+                None
+            )
             if self.config.enable_stateful_usage_ingestion:
                 redundant_usage_run_skip_handler = RedundantUsageRunSkipHandler(
                     source=self,
@@ -251,11 +250,6 @@ class SnowflakeV2Source(
 
         self.add_config_to_report()
 
-    @classmethod
-    def create(cls, config_dict: dict, ctx: PipelineContext) -> "Source":
-        config = SnowflakeV2Config.parse_obj(config_dict)
-        return cls(ctx, config)
-
     @staticmethod
     def test_connection(config_dict: dict) -> TestConnectionReport:
         test_report = TestConnectionReport()
@@ -302,7 +296,16 @@ class SnowflakeV2Source(
 
         _report: Dict[Union[SourceCapability, str], CapabilityReport] = dict()
         privileges: List[SnowflakePrivilege] = []
-        capabilities: List[SourceCapability] = [c.capability for c in SnowflakeV2Source.get_capabilities() if c.capability not in (SourceCapability.PLATFORM_INSTANCE, SourceCapability.DOMAINS, SourceCapability.DELETION_DETECTION)]  # type: ignore
+        capabilities: List[SourceCapability] = [
+            c.capability
+            for c in SnowflakeV2Source.get_capabilities()  # type: ignore
+            if c.capability
+            not in (
+                SourceCapability.PLATFORM_INSTANCE,
+                SourceCapability.DOMAINS,
+                SourceCapability.DELETION_DETECTION,
+            )
+        ]
 
         cur = conn.query("select current_role()")
         current_role = [row["CURRENT_ROLE()"] for row in cur][0]
@@ -480,8 +483,8 @@ class SnowflakeV2Source(
             identifiers=self.identifiers,
         )
 
-        self.report.set_ingestion_stage("*", METADATA_EXTRACTION)
-        yield from schema_extractor.get_workunits_internal()
+        with self.report.new_stage(f"*: {METADATA_EXTRACTION}"):
+            yield from schema_extractor.get_workunits_internal()
 
         databases = schema_extractor.databases
 
@@ -513,45 +516,46 @@ class SnowflakeV2Source(
         discovered_datasets = discovered_tables + discovered_views
 
         if self.config.use_queries_v2:
-            self.report.set_ingestion_stage("*", VIEW_PARSING)
-            yield from auto_workunit(self.aggregator.gen_metadata())
-
-            self.report.set_ingestion_stage("*", QUERIES_EXTRACTION)
-
-            schema_resolver = self.aggregator._schema_resolver
-
-            queries_extractor = SnowflakeQueriesExtractor(
-                connection=self.connection,
-                config=SnowflakeQueriesExtractorConfig(
-                    window=self.config,
-                    temporary_tables_pattern=self.config.temporary_tables_pattern,
-                    include_lineage=self.config.include_table_lineage,
-                    include_usage_statistics=self.config.include_usage_stats,
-                    include_operations=self.config.include_operational_stats,
-                    user_email_pattern=self.config.user_email_pattern,
-                ),
-                structured_report=self.report,
-                filters=self.filters,
-                identifiers=self.identifiers,
-                schema_resolver=schema_resolver,
-                discovered_tables=discovered_datasets,
-                graph=self.ctx.graph,
-            )
+            with self.report.new_stage(f"*: {VIEW_PARSING}"):
+                yield from auto_workunit(self.aggregator.gen_metadata())
 
-            # TODO: This is slightly suboptimal because we create two SqlParsingAggregator instances with different configs
-            # but a shared schema resolver. That's fine for now though - once we remove the old lineage/usage extractors,
-            # it should be pretty straightforward to refactor this and only initialize the aggregator once.
-            self.report.queries_extractor = queries_extractor.report
-            yield from queries_extractor.get_workunits_internal()
-            queries_extractor.close()
+            with self.report.new_stage(f"*: {QUERIES_EXTRACTION}"):
+                schema_resolver = self.aggregator._schema_resolver
+
+                queries_extractor = SnowflakeQueriesExtractor(
+                    connection=self.connection,
+                    config=SnowflakeQueriesExtractorConfig(
+                        window=self.config,
+                        temporary_tables_pattern=self.config.temporary_tables_pattern,
+                        include_lineage=self.config.include_table_lineage,
+                        include_usage_statistics=self.config.include_usage_stats,
+                        include_operations=self.config.include_operational_stats,
+                        include_queries=self.config.include_queries,
+                        include_query_usage_statistics=self.config.include_query_usage_statistics,
+                        user_email_pattern=self.config.user_email_pattern,
+                    ),
+                    structured_report=self.report,
+                    filters=self.filters,
+                    identifiers=self.identifiers,
+                    schema_resolver=schema_resolver,
+                    discovered_tables=discovered_datasets,
+                    graph=self.ctx.graph,
+                )
+
+                # TODO: This is slightly suboptimal because we create two SqlParsingAggregator instances with different configs
+                # but a shared schema resolver. That's fine for now though - once we remove the old lineage/usage extractors,
+                # it should be pretty straightforward to refactor this and only initialize the aggregator once.
+                self.report.queries_extractor = queries_extractor.report
+                yield from queries_extractor.get_workunits_internal()
+                queries_extractor.close()
 
         else:
             if self.lineage_extractor:
-                self.report.set_ingestion_stage("*", LINEAGE_EXTRACTION)
-                self.lineage_extractor.add_time_based_lineage_to_aggregator(
-                    discovered_tables=discovered_tables,
-                    discovered_views=discovered_views,
-                )
+                with self.report.new_stage(f"*: {LINEAGE_EXTRACTION}"):
+                    self.lineage_extractor.add_time_based_lineage_to_aggregator(
+                        discovered_tables=discovered_tables,
+                        discovered_views=discovered_views,
+                    )
 
             # This would emit view and external table ddl lineage
             # as well as query lineage via lineage_extractor

datahub/ingestion/source/sql/athena.py

@@ -104,9 +104,7 @@ class CustomAthenaRestDialect(AthenaRestDialect):
             return "\n".join([r for r in res])
 
     @typing.no_type_check
-    def _get_column_type(
-        self, type_: Union[str, Dict[str, Any]]
-    ) -> TypeEngine:  # noqa: C901
+    def _get_column_type(self, type_: Union[str, Dict[str, Any]]) -> TypeEngine:  # noqa: C901
         """Derives the data type of the Athena column.
 
         This method is overwritten to extend the behavior of PyAthena.

datahub/ingestion/source/sql/clickhouse.py

@@ -218,9 +218,7 @@ def _get_all_table_comments_and_properties(self, connection, **kw):
              , comment
              , {properties_clause} AS properties
           FROM system.tables
-         WHERE name NOT LIKE '.inner%'""".format(
-            properties_clause=properties_clause
-        )
+         WHERE name NOT LIKE '.inner%'""".format(properties_clause=properties_clause)
     )
 
     all_table_comments: Dict[Tuple[str, str], Dict[str, Any]] = {}
@@ -268,7 +266,7 @@ def _get_table_or_view_names(self, relkind, connection, schema=None, **kw):
     info_cache = kw.get("info_cache")
     all_relations = self._get_all_relation_info(connection, info_cache=info_cache)
     relation_names = []
-    for key, relation in all_relations.items():
+    for _, relation in all_relations.items():
         if relation.database == schema and relation.relkind == relkind:
             relation_names.append(relation.relname)
     return relation_names
@@ -301,9 +299,7 @@ def _get_schema_column_info(self, connection, schema=None, **kw):
              , comment
           FROM system.columns
          WHERE {schema_clause}
-         ORDER BY database, table, position""".format(
-                    schema_clause=schema_clause
-                )
+         ORDER BY database, table, position""".format(schema_clause=schema_clause)
             )
         )
     )
@@ -474,7 +470,7 @@ class ClickHouseSource(TwoTierSQLAlchemySource):
         logger.debug(f"sql_alchemy_url={url}")
         engine = create_engine(url, **self.config.options)
         for db_row in engine.execute(text(all_tables_query)):
-            all_tables_set.add(f'{db_row["database"]}.{db_row["table_name"]}')
+            all_tables_set.add(f"{db_row['database']}.{db_row['table_name']}")
 
         return all_tables_set
 
@@ -503,7 +499,7 @@ class ClickHouseSource(TwoTierSQLAlchemySource):
 
         try:
             for db_row in engine.execute(text(query)):
-                dataset_name = f'{db_row["target_schema"]}.{db_row["target_table"]}'
+                dataset_name = f"{db_row['target_schema']}.{db_row['target_table']}"
                 if not self.config.database_pattern.allowed(
                     db_row["target_schema"]
                 ) or not self.config.table_pattern.allowed(dataset_name):
@@ -512,7 +508,7 @@ class ClickHouseSource(TwoTierSQLAlchemySource):
 
                 # Target
                 target_path = (
-                    f'{self.config.platform_instance+"." if self.config.platform_instance else ""}'
+                    f"{self.config.platform_instance + '.' if self.config.platform_instance else ''}"
                     f"{dataset_name}"
                 )
                 target = LineageItem(
@@ -525,7 +521,7 @@ class ClickHouseSource(TwoTierSQLAlchemySource):
 
                 # Source
                 platform = LineageDatasetPlatform.CLICKHOUSE
-                path = f'{db_row["source_schema"]}.{db_row["source_table"]}'
+                path = f"{db_row['source_schema']}.{db_row['source_table']}"
 
                 sources = [
                     LineageDataset(
@@ -552,9 +548,7 @@ class ClickHouseSource(TwoTierSQLAlchemySource):
                         target.dataset.path
                     ].upstreams = self._lineage_map[
                         target.dataset.path
-                    ].upstreams.union(
-                        target.upstreams
-                    )
+                    ].upstreams.union(target.upstreams)
 
                 else:
                     self._lineage_map[target.dataset.path] = target

datahub/ingestion/source/sql/oracle.py

@@ -234,9 +234,7 @@ class OracleInspectorObjectWrapper:
                     WHERE col.table_name = id.table_name
                     AND col.column_name = id.column_name
                     AND col.owner = id.owner
-                ) AS identity_options""".format(
-                dblink=dblink
-            )
+                ) AS identity_options""".format(dblink=dblink)
         else:
             identity_cols = "NULL as default_on_null, NULL as identity_options"
 

datahub/ingestion/source/sql/sql_generic_profiler.py

@@ -278,8 +278,7 @@ class GenericProfiler:
 
         if self.config.profiling.profile_table_size_limit is not None and (
             size_in_bytes is not None
-            and size_in_bytes / (2**30)
-            > self.config.profiling.profile_table_size_limit
+            and size_in_bytes / (2**30) > self.config.profiling.profile_table_size_limit
         ):
             self.report.profiling_skipped_size_limit[schema_name] += 1
             logger.debug(

datahub/ingestion/source/sql/sql_types.py

@@ -93,7 +93,7 @@ POSTGRES_TYPES_MAP: Dict[str, Any] = {
     "regtype": None,
     "regrole": None,
     "regnamespace": None,
-    "super": None,
+    "super": NullType,
     "uuid": StringType,
     "pg_lsn": None,
     "tsvector": None,  # text search vector
@@ -384,7 +384,6 @@ TRINO_SQL_TYPES_MAP: Dict[str, Any] = {
     "varchar": StringType,
     "char": StringType,
     "varbinary": BytesType,
-    "json": RecordType,
     "date": DateType,
     "time": TimeType,
     "timestamp": TimeType,

datahub/ingestion/source/sql/sql_utils.py

@@ -20,6 +20,7 @@ from datahub.ingestion.api.workunit import MetadataWorkUnit
 from datahub.metadata.com.linkedin.pegasus2avro.dataset import UpstreamLineage
 from datahub.metadata.com.linkedin.pegasus2avro.schema import SchemaField
 from datahub.metadata.schema_classes import DataPlatformInstanceClass
+from datahub.metadata.urns import StructuredPropertyUrn
 from datahub.utilities.registries.domain_registry import DomainRegistry
 from datahub.utilities.urns.dataset_urn import DatasetUrn
 
@@ -75,6 +76,7 @@ def gen_schema_container(
     created: Optional[int] = None,
     last_modified: Optional[int] = None,
     extra_properties: Optional[Dict[str, str]] = None,
+    structured_properties: Optional[Dict[StructuredPropertyUrn, str]] = None,
 ) -> Iterable[MetadataWorkUnit]:
     domain_urn: Optional[str] = None
     if domain_registry:
@@ -99,6 +101,7 @@ def gen_schema_container(
         owner_urn=owner_urn,
         qualified_name=qualified_name,
         extra_properties=extra_properties,
+        structured_properties=structured_properties,
     )
 
 
@@ -133,6 +136,7 @@ def gen_database_container(
     created: Optional[int] = None,
     last_modified: Optional[int] = None,
     extra_properties: Optional[Dict[str, str]] = None,
+    structured_properties: Optional[Dict[StructuredPropertyUrn, str]] = None,
 ) -> Iterable[MetadataWorkUnit]:
     domain_urn: Optional[str] = None
     if domain_registry:
@@ -154,6 +158,7 @@ def gen_database_container(
         owner_urn=owner_urn,
         qualified_name=qualified_name,
         extra_properties=extra_properties,
+        structured_properties=structured_properties,
     )
 
 

datahub/ingestion/source/sql/teradata.py

@@ -599,7 +599,12 @@ ORDER by DataBaseName, TableName;
             setattr(  # noqa: B010
                 TeradataDialect,
                 "get_columns",
-                lambda self, connection, table_name, schema=None, use_qvci=self.config.use_qvci, **kw: optimized_get_columns(
+                lambda self,
+                connection,
+                table_name,
+                schema=None,
+                use_qvci=self.config.use_qvci,
+                **kw: optimized_get_columns(
                     self,
                     connection,
                     table_name,
@@ -613,7 +618,11 @@ ORDER by DataBaseName, TableName;
             setattr(  # noqa: B010
                 TeradataDialect,
                 "get_pk_constraint",
-                lambda self, connection, table_name, schema=None, **kw: optimized_get_pk_constraint(
+                lambda self,
+                connection,
+                table_name,
+                schema=None,
+                **kw: optimized_get_pk_constraint(
                     self, connection, table_name, schema, **kw
                 ),
             )
@@ -621,7 +630,11 @@ ORDER by DataBaseName, TableName;
             setattr(  # noqa: B010
                 TeradataDialect,
                 "get_foreign_keys",
-                lambda self, connection, table_name, schema=None, **kw: optimized_get_foreign_keys(
+                lambda self,
+                connection,
+                table_name,
+                schema=None,
+                **kw: optimized_get_foreign_keys(
                     self, connection, table_name, schema, **kw
                 ),
             )
@@ -878,7 +891,7 @@ ORDER by DataBaseName, TableName;
 
         urns = self.schema_resolver.get_urns()
         if self.config.include_table_lineage or self.config.include_usage_statistics:
-            self.report.report_ingestion_stage_start("audit log extraction")
-            yield from self.get_audit_log_mcps(urns=urns)
+            with self.report.new_stage("Audit log extraction"):
+                yield from self.get_audit_log_mcps(urns=urns)
 
         yield from self.builder.gen_workunits()

datahub/ingestion/source/state/profiling_state_handler.py

@@ -41,9 +41,9 @@ class ProfilingHandler(StatefulIngestionUsecaseHandlerBase[ProfilingCheckpointSt
         run_id: str,
     ):
         self.state_provider = source.state_provider
-        self.stateful_ingestion_config: Optional[
-            ProfilingStatefulIngestionConfig
-        ] = config.stateful_ingestion
+        self.stateful_ingestion_config: Optional[ProfilingStatefulIngestionConfig] = (
+            config.stateful_ingestion
+        )
         self.pipeline_name = pipeline_name
         self.run_id = run_id
         self.checkpointing_enabled: bool = (

datahub/ingestion/source/state/redundant_run_skip_handler.py

@@ -48,9 +48,9 @@ class RedundantRunSkipHandler(
     ):
         self.source = source
         self.state_provider = source.state_provider
-        self.stateful_ingestion_config: Optional[
-            StatefulIngestionConfig
-        ] = config.stateful_ingestion
+        self.stateful_ingestion_config: Optional[StatefulIngestionConfig] = (
+            config.stateful_ingestion
+        )
         self.pipeline_name = pipeline_name
         self.run_id = run_id
         self._job_id = self._init_job_id()
@@ -145,8 +145,7 @@ class RedundantRunSkipHandler(
             )
 
             logger.debug(
-                f"{self.job_id} : Last run start, end times:"
-                f"({last_run_time_window})"
+                f"{self.job_id} : Last run start, end times:({last_run_time_window})"
             )
 
             # If current run's time window is subset of last run's time window, then skip.
@@ -212,8 +211,7 @@ class RedundantRunSkipHandler(
             )
 
         self.log(
-            "Adjusted start, end times: "
-            f"({suggested_start_time}, {suggested_end_time})"
+            f"Adjusted start, end times: ({suggested_start_time}, {suggested_end_time})"
         )
         return (suggested_start_time, suggested_end_time)
 

datahub/ingestion/source/state/stale_entity_removal_handler.py

@@ -111,9 +111,9 @@ class StaleEntityRemovalHandler(
         self.state_type_class = state_type_class
         self.pipeline_name = pipeline_name
         self.run_id = run_id
-        self.stateful_ingestion_config: Optional[
-            StatefulStaleMetadataRemovalConfig
-        ] = config.stateful_ingestion
+        self.stateful_ingestion_config: Optional[StatefulStaleMetadataRemovalConfig] = (
+            config.stateful_ingestion
+        )
         self.checkpointing_enabled: bool = (
             True
             if (

datahub/ingestion/source/state_provider/datahub_ingestion_checkpointing_provider.py

@@ -70,20 +70,20 @@ class DatahubIngestionCheckpointingProvider(IngestionCheckpointingProviderBase):
             self.orchestrator_name, pipeline_name, job_name
         )
 
-        latest_checkpoint: Optional[
-            DatahubIngestionCheckpointClass
-        ] = self.graph.get_latest_timeseries_value(
-            entity_urn=data_job_urn,
-            aspect_type=DatahubIngestionCheckpointClass,
-            filter_criteria_map={
-                "pipelineName": pipeline_name,
-            },
+        latest_checkpoint: Optional[DatahubIngestionCheckpointClass] = (
+            self.graph.get_latest_timeseries_value(
+                entity_urn=data_job_urn,
+                aspect_type=DatahubIngestionCheckpointClass,
+                filter_criteria_map={
+                    "pipelineName": pipeline_name,
+                },
+            )
         )
         if latest_checkpoint:
             logger.debug(
                 f"The last committed ingestion checkpoint for pipelineName:'{pipeline_name}',"
                 f" job_name:'{job_name}' found with start_time:"
-                f" {datetime.utcfromtimestamp(latest_checkpoint.timestampMillis/1000)}"
+                f" {datetime.utcfromtimestamp(latest_checkpoint.timestampMillis / 1000)}"
             )
             return latest_checkpoint
         else:

datahub/ingestion/source/state_provider/file_ingestion_checkpointing_provider.py

@@ -67,7 +67,7 @@ class FileIngestionCheckpointingProvider(IngestionCheckpointingProviderBase):
             logger.debug(
                 f"The last committed ingestion checkpoint for pipelineName:'{pipeline_name}',"
                 f" job_name:'{job_name}' found with start_time:"
-                f" {datetime.utcfromtimestamp(latest_checkpoint.timestampMillis/1000)}"
+                f" {datetime.utcfromtimestamp(latest_checkpoint.timestampMillis / 1000)}"
             )
             return latest_checkpoint
         else:

datahub/ingestion/source/superset.py

@@ -33,7 +33,7 @@ from datahub.ingestion.api.decorators import (
     platform_name,
     support_status,
 )
-from datahub.ingestion.api.source import MetadataWorkUnitProcessor, Source
+from datahub.ingestion.api.source import MetadataWorkUnitProcessor
 from datahub.ingestion.api.workunit import MetadataWorkUnit
 from datahub.ingestion.source.sql.sql_types import resolve_sql_type
 from datahub.ingestion.source.sql.sqlalchemy_uri_mapper import (
@@ -265,11 +265,6 @@ class SupersetSource(StatefulIngestionSourceBase):
             # TODO(Gabe): how should we message about this error?
         return requests_session
 
-    @classmethod
-    def create(cls, config_dict: dict, ctx: PipelineContext) -> Source:
-        config = SupersetConfig.parse_obj(config_dict)
-        return cls(ctx, config)
-
     def paginate_entity_api_results(self, entity_type, page_size=100):
         current_page = 0
         total_items = page_size