openmetadata-ingestion 1.7.4.2__py3-none-any.whl → 1.8.0.0__py3-none-any.whl

metadata/ingestion/source/pipeline/nifi/metadata.py

@@ -12,13 +12,14 @@
 Nifi source to extract metadata
 """
 import traceback
-from typing import Iterable, List, Optional
+from collections import defaultdict
+from typing import Dict, Iterable, List, Optional
 
 from pydantic import BaseModel, ValidationError
 
 from metadata.generated.schema.api.data.createPipeline import CreatePipelineRequest
 from metadata.generated.schema.api.lineage.addLineage import AddLineageRequest
-from metadata.generated.schema.entity.data.pipeline import Task
+from metadata.generated.schema.entity.data.pipeline import Pipeline, Task
 from metadata.generated.schema.entity.services.connections.pipeline.nifiConnection import (
     NifiConnection,
 )
@@ -30,6 +31,9 @@ from metadata.generated.schema.type.basic import (
     FullyQualifiedEntityName,
     SourceUrl,
 )
+from metadata.generated.schema.type.entityLineage import EntitiesEdge, LineageDetails
+from metadata.generated.schema.type.entityLineage import Source as LineageSource
+from metadata.generated.schema.type.entityReference import EntityReference
 from metadata.ingestion.api.models import Either
 from metadata.ingestion.api.steps import InvalidSourceException
 from metadata.ingestion.models.pipeline_status import OMetaPipelineStatus
@@ -43,6 +47,7 @@ logger = ingestion_logger()
 
 PROCESS_GROUP_FLOW = "processGroupFlow"
 BREADCRUMB = "breadcrumb"
+PARENT_BREADCRUMB = "parentBreadcrumb"
 
 
 class NifiProcessor(BaseModel):
@@ -77,6 +82,7 @@ class NifiPipelineDetails(BaseModel):
     uri: str
     processors: List[NifiProcessor]
     connections: List[NifiProcessorConnections]
+    parent_pipeline_id: Optional[str] = None
 
 
 class NifiSource(PipelineServiceSource):
@@ -85,6 +91,11 @@ class NifiSource(PipelineServiceSource):
     Pipeline metadata from Airflow's metadata db
     """
 
+    def __init__(self, config: WorkflowSource, metadata: OpenMetadata):
+        super().__init__(config, metadata)
+        self.pipeline_parents_mapping: Dict[str, List[str]] = defaultdict(list)
+        self.process_group_connections: List[NifiProcessorConnections] = []
+
     @classmethod
     def create(
         cls, config_dict, metadata: OpenMetadata, pipeline_name: Optional[str] = None
@@ -220,7 +231,7 @@ class NifiSource(PipelineServiceSource):
         """Get List of all pipelines"""
         for process_group in self.connection.list_process_groups():
             try:
-                yield NifiPipelineDetails(
+                nifi_pipeline_details = NifiPipelineDetails(
                     id_=process_group[PROCESS_GROUP_FLOW].get("id"),
                     name=process_group[PROCESS_GROUP_FLOW][BREADCRUMB][BREADCRUMB].get(
                         "name"
@@ -232,7 +243,18 @@ class NifiSource(PipelineServiceSource):
                     connections=self._get_connections_from_process_group(
                         process_group=process_group
                     ),
+                    parent_pipeline_id=process_group[PROCESS_GROUP_FLOW][BREADCRUMB]
+                    .get(PARENT_BREADCRUMB, {})
+                    .get("id"),
+                )
+                if nifi_pipeline_details.parent_pipeline_id:
+                    self.pipeline_parents_mapping[nifi_pipeline_details.id_].append(
+                        nifi_pipeline_details.parent_pipeline_id
+                    )
+                self.process_group_connections.extend(
+                    self.get_process_group_connections(process_group)
                 )
+                yield nifi_pipeline_details
             except (ValueError, KeyError, ValidationError) as err:
                 logger.debug(traceback.format_exc())
                 logger.warning(
@@ -244,5 +266,113 @@ class NifiSource(PipelineServiceSource):
                     f"Wild error encountered when trying to get pipelines from Process Group {process_group} - {err}."
                 )
 
+    def get_process_group_connections(
+        self, process_group: dict
+    ) -> List[NifiProcessorConnections]:
+        """Get all connections for a process group"""
+        connections_list = (
+            process_group.get(PROCESS_GROUP_FLOW).get("flow").get("connections")
+        )
+        connections = []
+
+        for connection in connections_list:
+            try:
+                source = connection.get("component", {}).get("source", {})
+                destination = connection.get("component", {}).get("destination", {})
+                if (
+                    source.get("type") == "OUTPUT_PORT"
+                    and destination.get("type") == "INPUT_PORT"
+                    and source.get("groupId") != destination.get("groupId")
+                ):
+                    connections.append(
+                        NifiProcessorConnections(
+                            id_=connection.get("id"),
+                            source_id=source.get("groupId"),
+                            destination_id=destination.get("groupId"),
+                        )
+                    )
+            except Exception as err:
+                logger.debug(traceback.format_exc())
+                logger.warning(
+                    f"Wild error encountered when trying to get process group connections from \
+                    {process_group[PROCESS_GROUP_FLOW][BREADCRUMB][BREADCRUMB].get('name')} - {err}."
+                )
+        return connections
+
+    def yield_pipeline_bulk_lineage_details(
+        self,
+    ) -> Iterable[Either[AddLineageRequest]]:
+        """
+        Process the pipeline bulk lineage details
+        """
+        # Process the lineage between parent and child pipelines
+        for pipeline_id, parent_pipeline_ids in self.pipeline_parents_mapping.items():
+            to_entity = self.metadata.get_by_name(
+                entity=Pipeline,
+                fqn=f"{self.context.get().pipeline_service}.{pipeline_id}",
+            )
+            if not to_entity:
+                logger.warning(
+                    f"Pipeline {pipeline_id} not found in metadata, skipping lineage"
+                )
+                continue
+            for parent_pipeline_id in parent_pipeline_ids:
+                from_entity = self.metadata.get_by_name(
+                    entity=Pipeline,
+                    fqn=f"{self.context.get().pipeline_service}.{parent_pipeline_id}",
+                )
+                if not from_entity:
+                    logger.warning(
+                        f"Parent Pipeline {parent_pipeline_id} not found in metadata, skipping lineage"
+                    )
+                    continue
+                yield Either(
+                    right=AddLineageRequest(
+                        edge=EntitiesEdge(
+                            fromEntity=EntityReference(
+                                id=from_entity.id, type="pipeline"
+                            ),
+                            toEntity=EntityReference(id=to_entity.id, type="pipeline"),
+                            lineageDetails=LineageDetails(
+                                source=LineageSource.PipelineLineage
+                            ),
+                        )
+                    )
+                )
+
+        # Process the lineage between connected pipelines
+        for connection in self.process_group_connections:
+            from_entity = self.metadata.get_by_name(
+                entity=Pipeline,
+                fqn=f"{self.context.get().pipeline_service}.{connection.source_id}",
+            )
+            if not from_entity:
+                logger.warning(
+                    f"Pipeline {connection.source_id} not found in metadata, skipping lineage"
+                )
+                continue
+
+            to_entity = self.metadata.get_by_name(
+                entity=Pipeline,
+                fqn=f"{self.context.get().pipeline_service}.{connection.destination_id}",
+            )
+            if not to_entity:
+                logger.warning(
+                    f"Pipeline {connection.destination_id} not found in metadata, skipping lineage"
+                )
+                continue
+
+            yield Either(
+                right=AddLineageRequest(
+                    edge=EntitiesEdge(
+                        fromEntity=EntityReference(id=from_entity.id, type="pipeline"),
+                        toEntity=EntityReference(id=to_entity.id, type="pipeline"),
+                        lineageDetails=LineageDetails(
+                            source=LineageSource.PipelineLineage
+                        ),
+                    )
+                )
+            )
+
     def get_pipeline_name(self, pipeline_details: NifiPipelineDetails) -> str:
         return pipeline_details.name

metadata/ingestion/source/pipeline/pipeline_service.py

@@ -41,9 +41,6 @@ from metadata.ingestion.api.delete import delete_entity_from_source
 from metadata.ingestion.api.models import Either
 from metadata.ingestion.api.steps import Source
 from metadata.ingestion.api.topology_runner import TopologyRunnerMixin
-from metadata.ingestion.connections.test_connections import (
-    raise_test_connection_exception,
-)
 from metadata.ingestion.models.delete_entity import DeleteEntity
 from metadata.ingestion.models.ometa_classification import OMetaTagAndClassification
 from metadata.ingestion.models.ometa_lineage import OMetaLineageRequest
@@ -55,7 +52,7 @@ from metadata.ingestion.models.topology import (
     TopologyNode,
 )
 from metadata.ingestion.ometa.ometa_api import OpenMetadata
-from metadata.ingestion.source.connections import get_connection, get_test_connection_fn
+from metadata.ingestion.source.connections import get_connection, test_connection_common
 from metadata.ingestion.source.pipeline.openlineage.models import TableDetails
 from metadata.ingestion.source.pipeline.openlineage.utils import FQNNotFoundException
 from metadata.utils import fqn
@@ -137,6 +134,7 @@ class PipelineServiceTopology(ServiceTopology):
                 nullable=True,
             ),
         ],
+        post_process=["process_pipeline_bulk_lineage"],
     )
 
 
@@ -316,6 +314,23 @@ class PipelineServiceSource(TopologyRunnerMixin, Source, ABC):
                 else:
                     yield lineage
 
+    def yield_pipeline_bulk_lineage_details(self) -> Iterable[AddLineageRequest]:
+        """Method to yield the bulk pipeline lineage details"""
+
+    def process_pipeline_bulk_lineage(self) -> Iterable[AddLineageRequest]:
+        """Method to process the bulk pipeline lineage"""
+        if self.source_config.includeLineage:
+            for lineage in self.yield_pipeline_bulk_lineage_details() or []:
+                if lineage.right is not None:
+                    yield Either(
+                        right=OMetaLineageRequest(
+                            lineage_request=lineage.right,
+                            override_lineage=self.source_config.overrideLineage,
+                        )
+                    )
+                else:
+                    yield lineage
+
     def _get_table_fqn_from_om(self, table_details: TableDetails) -> Optional[str]:
         """
         Based on partial schema and table names look for matching table object in open metadata.
@@ -373,11 +388,9 @@ class PipelineServiceSource(TopologyRunnerMixin, Source, ABC):
             yield pipeline_detail
 
     def test_connection(self) -> None:
-        test_connection_fn = get_test_connection_fn(self.service_connection)
-        result = test_connection_fn(
+        test_connection_common(
             self.metadata, self.connection_obj, self.service_connection
         )
-        raise_test_connection_exception(result)
 
     def register_record(self, pipeline_request: CreatePipelineRequest) -> None:
         """Mark the pipeline record as scanned and update the pipeline_source_state"""

metadata/ingestion/source/search/search_service.py

@@ -41,9 +41,6 @@ from metadata.ingestion.api.delete import delete_entity_from_source
 from metadata.ingestion.api.models import Either
 from metadata.ingestion.api.steps import Source
 from metadata.ingestion.api.topology_runner import TopologyRunnerMixin
-from metadata.ingestion.connections.test_connections import (
-    raise_test_connection_exception,
-)
 from metadata.ingestion.models.delete_entity import DeleteEntity
 from metadata.ingestion.models.search_index_data import OMetaIndexSampleData
 from metadata.ingestion.models.topology import (
@@ -53,7 +50,7 @@ from metadata.ingestion.models.topology import (
     TopologyNode,
 )
 from metadata.ingestion.ometa.ometa_api import OpenMetadata
-from metadata.ingestion.source.connections import get_connection, get_test_connection_fn
+from metadata.ingestion.source.connections import get_connection, test_connection_common
 from metadata.utils import fqn
 from metadata.utils.filters import filter_by_search_index
 from metadata.utils.helpers import retry_with_docker_host
@@ -239,11 +236,9 @@ class SearchServiceSource(TopologyRunnerMixin, Source, ABC):
         """Nothing to prepare by default"""
 
     def test_connection(self) -> None:
-        test_connection_fn = get_test_connection_fn(self.service_connection)
-        result = test_connection_fn(
+        test_connection_common(
             self.metadata, self.connection_obj, self.service_connection
         )
-        raise_test_connection_exception(result)
 
     def mark_search_indexes_as_deleted(self) -> Iterable[Either[DeleteEntity]]:
         """Method to mark the search index as deleted"""

metadata/ingestion/source/storage/storage_service.py

@@ -40,9 +40,6 @@ from metadata.ingestion.api.delete import delete_entity_from_source
 from metadata.ingestion.api.models import Either
 from metadata.ingestion.api.steps import Source
 from metadata.ingestion.api.topology_runner import TopologyRunnerMixin
-from metadata.ingestion.connections.test_connections import (
-    raise_test_connection_exception,
-)
 from metadata.ingestion.models.delete_entity import DeleteEntity
 from metadata.ingestion.models.ometa_classification import OMetaTagAndClassification
 from metadata.ingestion.models.topology import (
@@ -52,7 +49,7 @@ from metadata.ingestion.models.topology import (
     TopologyNode,
 )
 from metadata.ingestion.ometa.ometa_api import OpenMetadata
-from metadata.ingestion.source.connections import get_connection, get_test_connection_fn
+from metadata.ingestion.source.connections import get_connection, test_connection_common
 from metadata.ingestion.source.database.glue.models import Column
 from metadata.readers.dataframe.models import DatalakeTableSchemaWrapper
 from metadata.readers.dataframe.reader_factory import SupportedTypes
@@ -240,11 +237,9 @@ class StorageServiceSource(TopologyRunnerMixin, Source, ABC):
         self.container_source_state.add(container_fqn)
 
     def test_connection(self) -> None:
-        test_connection_fn = get_test_connection_fn(self.service_connection)
-        result = test_connection_fn(
+        test_connection_common(
             self.metadata, self.connection_obj, self.service_connection
         )
-        raise_test_connection_exception(result)
 
     def mark_containers_as_deleted(self) -> Iterable[Either[DeleteEntity]]:
         """Method to mark the containers as deleted"""

metadata/profiler/interface/sqlalchemy/bigquery/profiler_interface.py

@@ -29,7 +29,7 @@ class BigQueryProfilerInterface(SQAProfilerInterface):
         from sqlalchemy_bigquery import STRUCT
 
         columns_list = []
-        for key, value in columns.items():
+        for key, value in columns:
             if not isinstance(value, STRUCT):
                 col = Column(f"{parent}.{key}", value)
                 # pylint: disable=protected-access
@@ -38,7 +38,7 @@ class BigQueryProfilerInterface(SQAProfilerInterface):
                 columns_list.append(col)
             else:
                 col = self._get_struct_columns(
-                    value.__dict__.get("_STRUCT_byname"), f"{parent}.{key}"
+                    value.__dict__.get("_STRUCT_fields"), f"{parent}.{key}"
                 )
                 columns_list.extend(col)
         return columns_list
@@ -53,7 +53,7 @@ class BigQueryProfilerInterface(SQAProfilerInterface):
             if isinstance(column.type, STRUCT):
                 columns.extend(
                     self._get_struct_columns(
-                        column.type.__dict__.get("_STRUCT_byname"), column.name
+                        column.type.__dict__.get("_STRUCT_fields"), column.name
                     )
                 )
             else:

metadata/profiler/metrics/static/not_regexp_match_count.py

@@ -18,6 +18,7 @@ from sqlalchemy import case, column, not_
 
 from metadata.generated.schema.configuration.profilerConfiguration import MetricType
 from metadata.profiler.metrics.core import StaticMetric, _label
+from metadata.profiler.orm.functions.regexp import RegexpMatchFn
 from metadata.profiler.orm.functions.sum import SumFn
 from metadata.profiler.orm.registry import is_concatenable
 
@@ -58,9 +59,9 @@ class NotRegexCount(StaticMetric):
                 [
                     (
                         not_(
-                            column(self.col.name, self.col.type).regexp_match(
-                                self.expression
-                            )
+                            RegexpMatchFn(
+                                column(self.col.name, self.col.type), self.expression
+                            ),
                         ),
                         0,
                     )

metadata/profiler/metrics/static/regexp_match_count.py

@@ -18,6 +18,7 @@ from sqlalchemy import case, column
 
 from metadata.generated.schema.configuration.profilerConfiguration import MetricType
 from metadata.profiler.metrics.core import StaticMetric, _label
+from metadata.profiler.orm.functions.regexp import RegexpMatchFn
 from metadata.profiler.orm.functions.sum import SumFn
 from metadata.profiler.orm.registry import is_concatenable
 
@@ -57,8 +58,8 @@ class RegexCount(StaticMetric):
             case(
                 [
                     (
-                        column(self.col.name, self.col.type).regexp_match(
-                            self.expression
+                        RegexpMatchFn(
+                            column(self.col.name, self.col.type), self.expression
                         ),
                         1,
                     )

metadata/profiler/orm/functions/length.py

@@ -52,6 +52,7 @@ def _(element, compiler, **kw):
 @compiles(LenFn, Dialects.Hana)
 @compiles(LenFn, Dialects.Druid)
 @compiles(LenFn, Dialects.Doris)
+@compiles(LenFn, Dialects.Teradata)
 def _(element, compiler, **kw):
     return "LENGTH(%s)" % compiler.process(element.clauses, **kw)
 

metadata/profiler/orm/functions/md5.py

@@ -40,3 +40,12 @@ def _(element, compiler, **kw):
 @compiles(MD5, PythonDialects.BigQuery.value)
 def _(element, compiler, **kw):
     return f"TO_HEX(MD5(CAST({compiler.process(element.clauses, **kw)} AS STRING)))"
+
+
+@compiles(MD5, PythonDialects.Teradata.value)
+def _(element, compiler, **kw):
+    # There is no MD5 in Teradata or any other hashes
+    # But we can use UDF function hash_md5 published by Teradata Community
+    return (
+        f"HASH_MD5(CAST({compiler.process(element.clauses, **kw)} AS VARCHAR(32000)))"
+    )

metadata/profiler/orm/functions/modulo.py

@@ -61,6 +61,7 @@ def _(element, compiler, **kw):
 @compiles(ModuloFn, Dialects.Vertica)
 @compiles(ModuloFn, Dialects.Hana)
 @compiles(ModuloFn, Dialects.Cockroach)
+@compiles(ModuloFn, Dialects.Teradata)
 def _(element, compiler, **kw):
     """Modulo function for specific dialect"""
     value, base = validate_and_compile(element, compiler, **kw)

metadata/profiler/orm/functions/random_num.py

@@ -98,10 +98,14 @@ def _(*_, **__):
 
 
 @compiles(RandomNumFn, Dialects.Snowflake)
+@compiles(RandomNumFn, Dialects.Teradata)
 def _(*_, **__):
     """We use FROM <table> SAMPLE BERNOULLI (n) for sampling
     in snowflake. We'll return 0 to make sure we get all the rows
     from the already sampled results when executing row::MOD(0, 100) < profile_sample.
+
+    Teradata RANDOM(0,100) function can't be used inside ORDER BY clause. That's why
+    use the same trick.
     """
     return "0"
 

metadata/profiler/orm/functions/regexp.py

@@ -0,0 +1,40 @@
+#  Copyright 2025 Collate
+#  Licensed under the Collate Community License, Version 1.0 (the "License");
+#  you may not use this file except in compliance with the License.
+#  You may obtain a copy of the License at
+#  https://github.com/open-metadata/OpenMetadata/blob/main/ingestion/LICENSE
+#  Unless required by applicable law or agreed to in writing, software
+#  distributed under the License is distributed on an "AS IS" BASIS,
+#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#  See the License for the specific language governing permissions and
+#  limitations under the License.
+
+"""Define a regexp match function."""
+
+from sqlalchemy.ext.compiler import compiles
+from sqlalchemy.sql.functions import GenericFunction
+
+from metadata.profiler.metrics.core import CACHE
+from metadata.profiler.orm.registry import Dialects
+
+
+class RegexpMatchFn(GenericFunction):
+    name = "regexp_match"
+    inherit_cache = CACHE
+
+
+@compiles(RegexpMatchFn)
+def _(element, compiler, **kw):
+    """Base function for regexp_match"""
+    column, pattern = element.clauses
+    fn = column.regexp_match(pattern)
+    return compiler.process(fn, **kw)
+
+
+@compiles(RegexpMatchFn, Dialects.Databricks)
+def _(element, compiler, **kw):
+    """Databricks function for regexp_match"""
+    column, pattern = element.clauses
+    compiled_column = compiler.process(column, **kw)
+    compiled_pattern = compiler.process(pattern, **kw)
+    return f"REGEXP_LIKE({compiled_column}, {compiled_pattern})"

metadata/profiler/orm/functions/table_metric_computer.py

@@ -16,7 +16,7 @@ Run profiler metrics on the table
 
 import traceback
 from abc import ABC, abstractmethod
-from typing import Callable, List, Optional, Tuple
+from typing import Callable, List, Optional, Tuple, Type
 
 from sqlalchemy import Column, MetaData, Table, func, inspect, literal, select
 from sqlalchemy.sql.expression import ColumnOperators, and_, cte
@@ -27,13 +27,29 @@ from metadata.generated.schema.entity.data.table import TableType
 from metadata.profiler.metrics.registry import Metrics
 from metadata.profiler.orm.registry import Dialects
 from metadata.profiler.processor.runner import QueryRunner
+from metadata.profiler.registry import MetricRegistry
+from metadata.utils.dependency_injector.dependency_injector import (
+    DependencyNotFoundError,
+    Inject,
+    inject,
+)
 from metadata.utils.logger import profiler_interface_registry_logger
 
 logger = profiler_interface_registry_logger()
 
+
+@inject
+def get_row_count_metric(metrics: Inject[Type[MetricRegistry]] = None):
+    if metrics is None:
+        raise DependencyNotFoundError(
+            "MetricRegistry dependency not found. Please ensure the MetricRegistry is properly registered."
+        )
+    return metrics.ROW_COUNT().name()
+
+
 COLUMN_COUNT = "columnCount"
 COLUMN_NAMES = "columnNames"
-ROW_COUNT = Metrics.ROW_COUNT().name()
+ROW_COUNT = get_row_count_metric()
 SIZE_IN_BYTES = "sizeInBytes"
 CREATE_DATETIME = "createDateTime"
 
@@ -336,12 +352,16 @@ class BigQueryTableMetricComputer(BaseTableMetricComputer):
             Column("dataset_id") == self.schema_name,
             Column("table_id") == self.table_name,
         ]
+        schema = (
+            self.schema_name.startswith(
+                f"{self.conn_config.credentials.gcpConfig.projectId.root}."
+            )
+            and self.schema_name
+            or f"{self.conn_config.credentials.gcpConfig.projectId.root}.{self.schema_name}"
+        )
         query = self._build_query(
             columns,
-            self._build_table(
-                "__TABLES__",
-                f"{self.conn_config.credentials.gcpConfig.projectId.root}.{self.schema_name}",
-            ),
+            self._build_table("__TABLES__", schema),
             where_clause,
         )
         res = self.runner._session.execute(query).first()
@@ -358,9 +378,15 @@ class BigQueryTableMetricComputer(BaseTableMetricComputer):
 class MySQLTableMetricComputer(BaseTableMetricComputer):
     """MySQL Table Metric Computer"""
 
-    def compute(self):
+    @inject
+    def compute(self, metrics: Inject[Type[MetricRegistry]] = None):
         """compute table metrics for mysql"""
 
+        if metrics is None:
+            raise DependencyNotFoundError(
+                "MetricRegistry dependency not found. Please ensure the MetricRegistry is properly registered."
+            )
+
         columns = [
             Column("TABLE_ROWS").label(ROW_COUNT),
             (Column("data_length") + Column("index_length")).label(SIZE_IN_BYTES),
@@ -386,7 +412,7 @@ class MySQLTableMetricComputer(BaseTableMetricComputer):
         res = res._asdict()
         # innodb row count is an estimate we need to patch the row count with COUNT(*)
         # https://dev.mysql.com/doc/refman/8.3/en/information-schema-innodb-tablestats-table.html
-        row_count = self.runner.select_first_from_table(Metrics.ROW_COUNT().fn())
+        row_count = self.runner.select_first_from_table(metrics.ROW_COUNT().fn())
         res.update({ROW_COUNT: row_count.rowCount})
         return res
 

metadata/profiler/orm/registry.py

@@ -86,6 +86,7 @@ class PythonDialects(Enum):
     SingleStore = "singlestore"
     SQLite = "sqlite"
     Snowflake = "snowflake"
+    Teradata = "teradatasql"
     Trino = "trino"
     Vertica = "vertica"
 

metadata/profiler/processor/core.py

@@ -16,7 +16,7 @@ from __future__ import annotations
 
 import traceback
 from datetime import datetime, timezone
-from typing import Any, Dict, Generic, List, Optional, Set, Tuple, Type
+from typing import Any, Dict, Generic, List, Optional, Set, Tuple, Type, cast
 
 from pydantic import ValidationError
 from sqlalchemy import Column
@@ -94,7 +94,7 @@ class Profiler(Generic[TMetric]):
         :param profile_sample: % of rows to use for sampling column metrics
         """
         self.global_profiler_configuration: Optional[ProfilerConfiguration] = (
-            global_profiler_configuration.config_value
+            cast(ProfilerConfiguration, global_profiler_configuration.config_value)
             if global_profiler_configuration
             else None
         )