arthur-common 1.0.1__py3-none-any.whl → 2.1.48__py3-none-any.whl

arthur_common/models/connectors.py

@@ -4,7 +4,6 @@ from pydantic import BaseModel, Field
 class ConnectorPaginationOptions(BaseModel):
     page: int = Field(default=1, ge=1)
     page_size: int = Field(default=25, gt=0, le=500)
-    # comment to run pipeline
 
     @property
     def page_params(self) -> tuple[int, int]:
@@ -30,6 +29,12 @@ GOOGLE_CONNECTOR_PROJECT_ID_FIELD = "project_id"
 GOOGLE_CONNECTOR_LOCATION_FIELD = "location"
 SHIELD_CONNECTOR_API_KEY_FIELD = "api_key"
 SHIELD_CONNECTOR_ENDPOINT_FIELD = "endpoint"
+MSSQL_CONNECTOR_HOST_FIELD = "host"                    
+MSSQL_CONNECTOR_PORT_FIELD = "port"
+MSSQL_CONNECTOR_DATABASE_FIELD = "database"
+MSSQL_CONNECTOR_USERNAME_FIELD = "username"
+MSSQL_CONNECTOR_PASSWORD_FIELD = "password"
+MSSQL_CONNECTOR_DRIVER_FIELD = "driver"
 
 # dataset (connector type dependent) constants
 SHIELD_DATASET_TASK_ID_FIELD = "task_id"

arthur_common/models/metrics.py

@@ -166,7 +166,7 @@ class MetricsColumnParameterSchema(MetricsParameterSchema):
 
 
 # Not used /implemented yet. Might turn into group by column list
-class MetricsColumnListParameterSchema(MetricsParameterSchema):
+class MetricsColumnListParameterSchema(MetricsColumnParameterSchema):
     parameter_type: Literal["column_list"] = "column_list"
 
 
@@ -202,13 +202,6 @@ class AggregationSpecSchema(BaseModel):
         description="List of parameters to the aggregation's aggregate function.",
     )
 
-    def parameter_is_column_reference(self, parameter_name: str) -> bool:
-        return any(
-            param.parameter_key == parameter_name
-            and isinstance(param, MetricsColumnParameterSchema)
-            for param in self.aggregate_args
-        )
-
     @model_validator(mode="after")
     def column_dataset_references_exist(self) -> Self:
         dataset_parameter_keys = [
@@ -218,7 +211,10 @@ class AggregationSpecSchema(BaseModel):
         ]
         for param in self.aggregate_args:
             if (
-                isinstance(param, MetricsColumnParameterSchema)
+                isinstance(
+                    param,
+                    (MetricsColumnParameterSchema, MetricsColumnListParameterSchema),
+                )
                 and param.source_dataset_parameter_key not in dataset_parameter_keys
             ):
                 raise ValueError(

arthur_common/models/schema_definitions.py

@@ -17,6 +17,7 @@ class ScopeSchemaTag(str, Enum):
     CONTINUOUS = "continuous"
     PREDICTION = "prediction"
     GROUND_TRUTH = "ground_truth"
+    PIN_IN_DEEP_DIVE = "pin_in_deep_dive"
 
 
 class DType(str, Enum):
@@ -29,6 +30,7 @@ class DType(str, Enum):
     TIMESTAMP = "timestamp"
     DATE = "date"
     JSON = "json"
+    IMAGE = "image"
 
 
 class MetricParameterAnnotation(BaseModel):

arthur_common/tools/aggregation_analyzer.py

@@ -13,6 +13,7 @@ from arthur_common.models.metrics import (
     AggregationMetricType,
     AggregationSpecSchema,
     DatasetReference,
+    MetricsColumnListParameterSchema,
     MetricsColumnParameterSchema,
     MetricsDatasetParameterSchema,
     MetricsLiteralParameterSchema,
@@ -23,6 +24,7 @@ from arthur_common.models.schema_definitions import (
     MetricColumnParameterAnnotation,
     MetricDatasetParameterAnnotation,
     MetricLiteralParameterAnnotation,
+    MetricMultipleColumnParameterAnnotation,
     MetricsParameterAnnotationUnion,
 )
 
@@ -44,6 +46,19 @@ class FunctionAnalyzer:
             return DType.UUID
         elif t is DatasetReference:
             return DType.UUID
+        elif typing.get_origin(t) is list:
+            return DType.JSON
+        elif typing.get_origin(t) is typing.Union:
+            # handle union types to add support for Optional types only
+            # extract the non-None type from Optional[T] = Union[T, None]
+            args = typing.get_args(t)
+            non_none_args = [arg for arg in args if arg is not type(None)]
+            if len(non_none_args) == 1:
+                return FunctionAnalyzer._python_type_to_scope_dtype(non_none_args[0])
+            else:
+                raise ValueError(
+                    f"Union type {t} is not supported (only Optional[T] is supported).",
+                )
         else:
             raise ValueError(f"Parameter type {t} is not supported.")
 
@@ -69,7 +84,7 @@ class FunctionAnalyzer:
     @staticmethod
     def _get_scope_metric_parameter_from_annotation(
         param_name: str,
-        param_dtype: DType,
+        param_dtype: typing.Optional[DType],
         optional: bool,
         annotation: typing.Annotated,  # type: ignore
     ) -> MetricsParameterSchemaUnion:
@@ -101,6 +116,21 @@ class FunctionAnalyzer:
                 description=annotation.description,
                 model_problem_type=annotation.model_problem_type,
             )
+        elif isinstance(annotation, MetricMultipleColumnParameterAnnotation):
+            if param_dtype != DType.JSON:
+                raise ValueError(
+                    f"Dataset parameter {param_name} has type {param_dtype}, but should be a JSON type (valid list expected).",
+                )
+            return MetricsColumnListParameterSchema(
+                parameter_key=param_name,
+                tag_hints=annotation.tag_hints,
+                optional=optional,
+                source_dataset_parameter_key=annotation.source_dataset_parameter_key,
+                allowed_column_types=annotation.allowed_column_types,
+                allow_any_column_type=annotation.allow_any_column_type,
+                friendly_name=annotation.friendly_name,
+                description=annotation.description,
+            )
         elif isinstance(annotation, MetricColumnParameterAnnotation):
             if param_dtype != DType.STRING:
                 raise ValueError(

arthur_common/tools/duckdb_data_loader.py

@@ -291,7 +291,7 @@ def _make_schema(
                 return "DOUBLE"
             case DType.BOOL:
                 return "BOOLEAN"
-            case DType.STRING:
+            case DType.STRING | DType.IMAGE:
                 return "VARCHAR"
             case DType.UUID:
                 return "UUID"

{arthur_common-1.0.1.dist-info → arthur_common-2.1.48.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: arthur-common
-Version: 1.0.1
+Version: 2.1.48
 Summary: Utility code common to Arthur platform components.
 License: MIT
 Author: Arthur
@@ -69,6 +69,3 @@ This project is licensed under the MIT License.
 
 - Arthur <engineering@arthur.ai>
 
-# ALEX
-- Change for testing
-

{arthur_common-1.0.1.dist-info → arthur_common-2.1.48.dist-info}/RECORD

@@ -1,40 +1,39 @@
 arthur_common/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-arthur_common/__version__.py,sha256=d4QHYmS_30j0hPN8NmNPnQ_Z0TphDRbu4MtQj9cT9e8,22
 arthur_common/aggregations/__init__.py,sha256=vISWyciQAtksa71OKeHNP-QyFGd1NzBKq_LBsG0QSG8,67
-arthur_common/aggregations/aggregator.py,sha256=cw6mr1Dl0sH7Rn_EggCRvduC7GvC8XTDP2L84FODKY0,7445
+arthur_common/aggregations/aggregator.py,sha256=8J0TRYjQdwf0qXBnoZ4ETJVCdo4UIyvQnBTZCypY7Fo,7713
 arthur_common/aggregations/functions/README.md,sha256=MkZoTAJ94My96R5Z8GAxud7S6vyR0vgVi9gqdt9a4XY,5460
 arthur_common/aggregations/functions/__init__.py,sha256=HqC3UNRURX7ZQHgamTrQvfA8u_FiZGZ4I4eQW7Ooe5o,1299
-arthur_common/aggregations/functions/categorical_count.py,sha256=_hk1TYkrnCAe-0Hflt4W3Nvrp-c-vWHSgVxaqcFNWAI,3368
-arthur_common/aggregations/functions/confusion_matrix.py,sha256=xqJkwcFZck_F1tsvMeJdmU-EOSOfWJS_N7-PXYJiSmo,16351
-arthur_common/aggregations/functions/inference_count.py,sha256=BGWa262UxpkqY5Y_Pm22URSRvwXzdRbGUcd4DSVJbN0,2312
-arthur_common/aggregations/functions/inference_count_by_class.py,sha256=BwZw8wCmFHRoau5oDgcDAkcwNhXjktFBCG1Sfsj6iGY,7830
-arthur_common/aggregations/functions/inference_null_count.py,sha256=wZDz89_23bGjB_Tb3ob_69_VhugFxzbSRkuENfyJ-ic,2867
-arthur_common/aggregations/functions/mean_absolute_error.py,sha256=rDEQlKEDyy_zRewtgthB_BK2oKrW6ymWmdtkxUFzfW8,4233
-arthur_common/aggregations/functions/mean_squared_error.py,sha256=TMUyPPPEHuG9QFmD2gZxmxy-f_CDU4ds_2R5-DFr42c,4251
-arthur_common/aggregations/functions/multiclass_confusion_matrix.py,sha256=QM27MhZMvF5Q5yxENSsHx_MMV_5h6eiRPP554jsdBqY,8204
-arthur_common/aggregations/functions/multiclass_inference_count_by_class.py,sha256=dmf4OI4KcFVvE61S2TnPZvBlNOhlFNFKrV7x2_ys5ZU,3201
-arthur_common/aggregations/functions/numeric_stats.py,sha256=eN1q10KquF8GeBOOktBuatV9Zxvd_--ZzklqzDpT9qw,3124
-arthur_common/aggregations/functions/numeric_sum.py,sha256=DO7jbmUEnyyJliEtYLHb58SDCVBTsFKvywFnADXFpw8,3036
+arthur_common/aggregations/functions/categorical_count.py,sha256=W0xIpqqSElRpGBPpvztoZBpDO5QtiGYpaMVOQ3piS-4,5077
+arthur_common/aggregations/functions/confusion_matrix.py,sha256=fZ5SeFcENjGyoCqecRTF9Y0Ub-RzsA-SlZ8cEHzUvnM,21111
+arthur_common/aggregations/functions/inference_count.py,sha256=jIhuh_NjL82NA84qJD2VI3FVRgZCSFjfcME3Hr_slpg,3832
+arthur_common/aggregations/functions/inference_count_by_class.py,sha256=JKTOhXSP3hBz28u10AvA8OyDi7P68Fv7NP6j_M7kXCQ,11066
+arthur_common/aggregations/functions/inference_null_count.py,sha256=l-yuVX7OJQGk-vvnkuXJPD-mEI1ZzwGjbzeihwSDT_M,4566
+arthur_common/aggregations/functions/mean_absolute_error.py,sha256=pgputW69w0DUT4xbj2nfnPpqliHZAtvh7TQnBfPQ584,6000
+arthur_common/aggregations/functions/mean_squared_error.py,sha256=o9TTdlBb1lX57kWqoMRdTNacXcEa_VH-AkfT6SL1xGs,6032
+arthur_common/aggregations/functions/multiclass_confusion_matrix.py,sha256=47CVl-Wo40jDeKvSKHeWB2DFFTDo_ichXPGyMmlVEmo,10973
+arthur_common/aggregations/functions/multiclass_inference_count_by_class.py,sha256=gLdjXKvQ_WDscc3zksH92LGHhuknkjf6oAF63-cwbJo,3978
+arthur_common/aggregations/functions/numeric_stats.py,sha256=9zyroFSYXXELnz4MKow8v06-tPVL23COVafoLHaHgVc,4676
+arthur_common/aggregations/functions/numeric_sum.py,sha256=f0ZIYqPjBsyCh-V8ktHpDIs1_4wjjPENboC607i5O3c,4785
 arthur_common/aggregations/functions/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 arthur_common/aggregations/functions/shield_aggregations.py,sha256=nkMlj9V7NIKeRP46jpsrlfB741RHk2CgwimD7BYp9To,31540
 arthur_common/aggregations/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 arthur_common/models/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-arthur_common/models/connectors.py,sha256=Ng2hwXQumfLE9SYepYZs2L5Y6aAIXAQfkJIa0rKCvWQ,1565
+arthur_common/models/connectors.py,sha256=hj8epo32MjtDgWGon1U4l__Ls6pf8nFp2v_jrUYUTCI,1799
 arthur_common/models/datasets.py,sha256=giG_8mv_3ilBf7cIvRV0_TDCDdb4qxRbYZvl7hRb6l8,491
-arthur_common/models/metrics.py,sha256=gFEGuM4kuac2CqpPN69BM2cWG-SUPE4-1jZVHv_M3M0,8380
+arthur_common/models/metrics.py,sha256=ZfsDL3aPdhZ1-E1-wlRyZT4lZNzz2A3qL-jDrqJbQdY,8217
 arthur_common/models/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-arthur_common/models/schema_definitions.py,sha256=MiM-oynqh71zZvVyM-XF5nN55dgOeqEajVyL8ZE3Wuo,14571
+arthur_common/models/schema_definitions.py,sha256=WBJ6JvBCvThlMBGQsbiAwi9QYjhG9KyURyjE8an8-yE,14633
 arthur_common/models/shield.py,sha256=1ZblfULKCf5BEvYURO5WScyfmijGwjAmcj0XADlF-XY,19110
 arthur_common/models/task_job_specs.py,sha256=GLJ7qmrb5eXnl5PiV27nnx_yG4S4sc4NDJ8-6xmNDLM,2796
 arthur_common/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 arthur_common/tools/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-arthur_common/tools/aggregation_analyzer.py,sha256=e4F8vsYDRRTzUmNVIl1vFrn9_nEeYDYcP3ygk7i1964,9534
+arthur_common/tools/aggregation_analyzer.py,sha256=IOpGfj-aihjUNo_L5iI47x2EuJWXAW0FKOUMzNca7ek,11130
 arthur_common/tools/aggregation_loader.py,sha256=3CF46bNi-GdJBNOXkjYfCQ1Aung8lf65L532sdWmR_s,2351
-arthur_common/tools/duckdb_data_loader.py,sha256=XrdXRFkgiGtYulOGsC4khVf12sNiSFx5hB5vD7vQzFE,11066
+arthur_common/tools/duckdb_data_loader.py,sha256=ywvFpI1qioK5Is-S7XxBqgQmQNY6u21qibKygTAW0Oo,11080
 arthur_common/tools/functions.py,sha256=FWL4eWO5-vLp86WudT-MGUKvf2B8f02IdoXQFKd6d8k,1093
 arthur_common/tools/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 arthur_common/tools/schema_inferer.py,sha256=PkAOHZRk_rZ1OZSigYrfzH-jERb9B_Gu7pOMl9WJQA8,4202
 arthur_common/tools/time_utils.py,sha256=4gfiu9NXfvPZltiVNLSIQGylX6h2W0viNi9Kv4bKyfw,1410
-arthur_common-1.0.1.dist-info/METADATA,sha256=WrMAk42ZfTrZhAz12r127fJhuglMx-zHnDXPMp7tzsk,1596
-arthur_common-1.0.1.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
-arthur_common-1.0.1.dist-info/RECORD,,
+arthur_common-2.1.48.dist-info/METADATA,sha256=seHFqwRDe2hVB1k-mNZRKd2ZgdOMNAhJjfzh8srCkk8,1568
+arthur_common-2.1.48.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
+arthur_common-2.1.48.dist-info/RECORD,,

arthur_common/__version__.py

@@ -1 +0,0 @@
-__version__ = "1.0.1"