refactor(pydantic): migrate all remaining 'class Config' to Pydantic v2 model_config

- Migrate ConfigModel, PermissiveConfigModel, and ConnectionModel to use ConfigDict
- Convert 16 more class Config instances across 13 files to model_config = ConfigDict(...)
- Remove all Pydantic v1 'class Config:' syntax from the codebase
- Fixes all Pydantic v2 deprecation warnings related to class-based config

Files updated:
- common.py: Base ConfigModel, PermissiveConfigModel, ConnectionModel
- sql_queries.py: QueryEvent model
- datahub_classifier.py: PredictionFactorsAndWeights, InfoTypeConfig
- schema_assertion.py: FieldListSchemaContract
- datahub_secret_store.py: DataHubSecretStoreConfig
- add_dataset_properties.py: AddDatasetPropertiesConfig
- superset.py: SupersetConfig
- snowflake_usage_v2.py: PermissiveModel
- report_server_domain.py: OwnershipData
- hex/api.py: HexApiSharing, HexApiProjectApiResource, HexApiProjectsListResponse
- dbt_core.py: DBTRunResult
- path_spec.py: PathSpec
- graph/config.py: DatahubClientConfig
- serialized_value.py: SerializedResourceValue

Author: sgomezvillamor

metadata-ingestion/src/datahub/api/entities/common/serialized_value.py

@@ -3,7 +3,7 @@
 from typing import Dict, Optional, Type, TypeVar, Union
 
 from avrogen.dict_wrapper import DictWrapper
-from pydantic import BaseModel
+from pydantic import BaseModel, ConfigDict
 
 import datahub.metadata.schema_classes as models
 from datahub.metadata.schema_classes import __SCHEMA_TYPES as SCHEMA_TYPES
@@ -17,8 +17,7 @@
 
 
 class SerializedResourceValue(BaseModel):
-    class Config:
-        arbitrary_types_allowed = True
+    model_config = ConfigDict(arbitrary_types_allowed=True)
 
     content_type: str
     blob: bytes

metadata-ingestion/src/datahub/api/entities/datacontract/schema_assertion.py

@@ -3,7 +3,7 @@
 import json
 from typing import List, Union
 
-from pydantic import Field, RootModel
+from pydantic import ConfigDict, Field, RootModel
 from typing_extensions import Literal
 
 from datahub.api.entities.datacontract.assertion import BaseAssertion
@@ -36,8 +36,7 @@ def model_post_init(self, __context: object) -> None:
 
 
 class FieldListSchemaContract(BaseAssertion):
-    class Config:
-        arbitrary_types_allowed = True
+    model_config = ConfigDict(arbitrary_types_allowed=True)
 
     type: Literal["field-list"]
 

metadata-ingestion/src/datahub/configuration/common.py

@@ -155,15 +155,13 @@ class PermissiveConfigModel(ConfigModel):
     # but still allow the user to pass in arbitrary fields that we don't care about.
     # It is usually used for argument bags that are passed through to third-party libraries.
 
-    class Config:
-        extra = "allow"
+    model_config = ConfigDict(extra="allow")
 
 
 class ConnectionModel(BaseModel):
     """Represents the config associated with a connection"""
 
-    class Config:
-        extra = "allow"
+    model_config = ConfigDict(extra="allow")
 
 
 class TransformerSemantics(ConfigEnum):

metadata-ingestion/src/datahub/ingestion/glossary/datahub_classifier.py

@@ -3,7 +3,7 @@
 from datahub_classify.helper_classes import ColumnInfo
 from datahub_classify.infotype_predictor import predict_infotypes
 from datahub_classify.reference_input import input1 as default_config
-from pydantic import field_validator
+from pydantic import ConfigDict, field_validator
 from pydantic.fields import Field
 
 from datahub.configuration.common import ConfigModel
@@ -49,8 +49,7 @@ class ValuesFactorConfig(ConfigModel):
 
 
 class PredictionFactorsAndWeights(ConfigModel):
-    class Config:
-        populate_by_name = True
+    model_config = ConfigDict(populate_by_name=True)
 
     Name: float = Field(alias="name")
     Description: float = Field(alias="description")
@@ -59,8 +58,7 @@ class Config:
 
 
 class InfoTypeConfig(ConfigModel):
-    class Config:
-        populate_by_name = True
+    model_config = ConfigDict(populate_by_name=True)
 
     Prediction_Factors_and_Weights: PredictionFactorsAndWeights = Field(
         description="Factors and their weights to consider when predicting info types",

metadata-ingestion/src/datahub/ingestion/graph/config.py

@@ -1,6 +1,8 @@
 from enum import Enum, auto
 from typing import Dict, List, Optional
 
+from pydantic import ConfigDict
+
 from datahub.configuration.common import ConfigModel
 from datahub.configuration.env_vars import get_datahub_component
 
@@ -31,5 +33,4 @@ class DatahubClientConfig(ConfigModel):
     datahub_component: Optional[str] = None
     server_config_refresh_interval: Optional[int] = None
 
-    class Config:
-        extra = "ignore"
+    model_config = ConfigDict(extra="ignore")

metadata-ingestion/src/datahub/ingestion/source/data_lake_common/path_spec.py

@@ -7,7 +7,7 @@
 
 import parse
 from cached_property import cached_property
-from pydantic import field_validator, model_validator
+from pydantic import ConfigDict, field_validator, model_validator
 from pydantic.fields import Field
 from wcmatch import pathlib
 
@@ -83,8 +83,7 @@ class FolderTraversalMethod(Enum):
 
 
 class PathSpec(ConfigModel):
-    class Config:
-        arbitrary_types_allowed = True
+    model_config = ConfigDict(arbitrary_types_allowed=True)
 
     include: str = Field(
         description="Path to table. Name variable `{table}` is used to mark the folder with dataset. In absence of `{table}`, file level dataset will be created. Check below examples for more details.",

metadata-ingestion/src/datahub/ingestion/source/dbt/dbt_core.py

@@ -9,7 +9,7 @@
 import dateutil.parser
 import requests
 from packaging import version
-from pydantic import BaseModel, Field, model_validator
+from pydantic import BaseModel, ConfigDict, Field, model_validator
 
 from datahub.configuration.git import GitReference
 from datahub.configuration.validate_field_rename import pydantic_renamed_field
@@ -341,8 +341,7 @@ class DBTRunTiming(BaseModel):
 
 
 class DBTRunResult(BaseModel):
-    class Config:
-        extra = "allow"
+    model_config = ConfigDict(extra="allow")
 
     status: str
     timing: List[DBTRunTiming] = []

metadata-ingestion/src/datahub/ingestion/source/hex/api.py

@@ -4,7 +4,7 @@
 from typing import Any, Dict, Generator, List, Optional, Union
 
 import requests
-from pydantic import BaseModel, Field, ValidationError, field_validator
+from pydantic import BaseModel, ConfigDict, Field, ValidationError, field_validator
 from requests.adapters import HTTPAdapter
 from typing_extensions import assert_never
 from urllib3.util.retry import Retry
@@ -133,8 +133,7 @@ class HexApiSharing(BaseModel):
     collections: Optional[List[HexApiCollectionAccess]] = []
     groups: Optional[List[Any]] = []
 
-    class Config:
-        extra = "ignore"  # Allow extra fields in the JSON
+    model_config = ConfigDict(extra="ignore")  # Allow extra fields in the JSON
 
 
 class HexApiItemType(StrEnum):
@@ -165,8 +164,7 @@ class HexApiProjectApiResource(BaseModel):
     schedules: Optional[List[HexApiSchedule]] = []
     sharing: Optional[HexApiSharing] = None
 
-    class Config:
-        extra = "ignore"  # Allow extra fields in the JSON
+    model_config = ConfigDict(extra="ignore")  # Allow extra fields in the JSON
 
     @field_validator(
         "created_at",
@@ -200,8 +198,7 @@ class HexApiProjectsListResponse(BaseModel):
     values: List[HexApiProjectApiResource]
     pagination: Optional[HexApiPageCursors] = None
 
-    class Config:
-        extra = "ignore"  # Allow extra fields in the JSON
+    model_config = ConfigDict(extra="ignore")  # Allow extra fields in the JSON
 
 
 @dataclass

metadata-ingestion/src/datahub/ingestion/source/powerbi_report_server/report_server_domain.py

@@ -1,7 +1,7 @@
 from datetime import datetime
 from typing import Any, Dict, List, Optional
 
-from pydantic import BaseModel, Field, model_validator
+from pydantic import BaseModel, ConfigDict, Field, model_validator
 
 from datahub.ingestion.source.powerbi_report_server.constants import (
     RelationshipDirection,
@@ -357,8 +357,7 @@ class OwnershipData(BaseModel):
     existing_owners: Optional[List[OwnerClass]] = []
     owner_to_add: Optional[CorpUser] = None
 
-    class Config:
-        arbitrary_types_allowed = True
+    model_config = ConfigDict(arbitrary_types_allowed=True)
 
 
 CatalogItem.update_forward_refs()

metadata-ingestion/src/datahub/ingestion/source/snowflake/snowflake_usage_v2.py

@@ -5,6 +5,7 @@
 from typing import Any, Dict, Iterable, List, Optional, Tuple
 
 import pydantic
+from pydantic import ConfigDict
 
 from datahub.configuration.time_window_config import BaseTimeWindowConfig
 from datahub.emitter.mce_builder import make_user_urn
@@ -70,8 +71,7 @@
 
 
 class PermissiveModel(pydantic.BaseModel):
-    class Config:
-        extra = "allow"
+    model_config = ConfigDict(extra="allow")
 
 
 class SnowflakeColumnReference(PermissiveModel):