fix: Remove date partition column field from datasources that don't s… (#2478)

* fix: Remove date partition column field from datasources that don't support them

Signed-off-by: Achal Shah <[email protected]>

* remove for file sources as well

Signed-off-by: Achal Shah <[email protected]>

* remove other references

Signed-off-by: Achal Shah <[email protected]>

* remove other references

Signed-off-by: Achal Shah <[email protected]>

* reference to removal

Signed-off-by: Achal Shah <[email protected]>

* reference to removal

Signed-off-by: Achal Shah <[email protected]>

* reorder

Signed-off-by: Achal Shah <[email protected]>

* remove more:

Signed-off-by: Achal Shah <[email protected]>

* docs

Signed-off-by: Achal Shah <[email protected]>

* comment

Signed-off-by: Achal Shah <[email protected]>

* cr

Signed-off-by: Achal Shah <[email protected]>

Author: achals

sdk/python/feast/infra/offline_stores/bigquery_source.py

@@ -21,7 +21,7 @@ def __init__(
         table_ref: Optional[str] = None,
         created_timestamp_column: Optional[str] = "",
         field_mapping: Optional[Dict[str, str]] = None,
-        date_partition_column: Optional[str] = "",
+        date_partition_column: Optional[str] = None,
         query: Optional[str] = None,
         name: Optional[str] = None,
         description: Optional[str] = "",
@@ -37,7 +37,7 @@ def __init__(
             created_timestamp_column (optional): Timestamp column when row was created, used for deduplicating rows.
             field_mapping: A dictionary mapping of column names in this data source to feature names in a feature table
                 or view. Only used for feature columns, not entities or timestamp columns.
-            date_partition_column (optional): Timestamp column used for partitioning.
+            date_partition_column (deprecated): Timestamp column used for partitioning.
             query (optional): SQL query to execute to generate data for this data source.
             name (optional): Name for the source. Defaults to the table_ref if not specified.
             description (optional): A human-readable description.
@@ -61,6 +61,15 @@ def __init__(
             table = table_ref
         self.bigquery_options = BigQueryOptions(table_ref=table, query=query)
 
+        if date_partition_column:
+            warnings.warn(
+                (
+                    "The argument 'date_partition_column' is not supported for BigQuery sources. "
+                    "It will be removed in Feast 0.21+"
+                ),
+                DeprecationWarning,
+            )
+
         # If no name, use the table_ref as the default name
         _name = name
         if not _name:
@@ -78,10 +87,9 @@ def __init__(
 
         super().__init__(
             _name if _name else "",
-            event_timestamp_column,
-            created_timestamp_column,
-            field_mapping,
-            date_partition_column,
+            event_timestamp_column=event_timestamp_column,
+            created_timestamp_column=created_timestamp_column,
+            field_mapping=field_mapping,
             description=description,
             tags=tags,
             owner=owner,
@@ -128,7 +136,6 @@ def from_proto(data_source: DataSourceProto):
             table_ref=data_source.bigquery_options.table_ref,
             event_timestamp_column=data_source.event_timestamp_column,
             created_timestamp_column=data_source.created_timestamp_column,
-            date_partition_column=data_source.date_partition_column,
             query=data_source.bigquery_options.query,
             description=data_source.description,
             tags=dict(data_source.tags),
@@ -148,7 +155,6 @@ def to_proto(self) -> DataSourceProto:
 
         data_source_proto.event_timestamp_column = self.event_timestamp_column
         data_source_proto.created_timestamp_column = self.created_timestamp_column
-        data_source_proto.date_partition_column = self.date_partition_column
 
         return data_source_proto
 

sdk/python/feast/infra/offline_stores/contrib/spark_offline_store/spark_source.py

@@ -51,12 +51,21 @@ def __init__(
                 _name = table
             else:
                 raise DataSourceNoNameException()
+
+        if date_partition_column:
+            warnings.warn(
+                (
+                    "The argument 'date_partition_column' is not supported for Spark sources."
+                    "It will be removed in Feast 0.21+"
+                ),
+                DeprecationWarning,
+            )
+
         super().__init__(
-            _name,
-            event_timestamp_column,
-            created_timestamp_column,
-            field_mapping,
-            date_partition_column,
+            name=_name,
+            event_timestamp_column=event_timestamp_column,
+            created_timestamp_column=created_timestamp_column,
+            field_mapping=field_mapping,
             description=description,
             tags=tags,
             owner=owner,
@@ -130,7 +139,6 @@ def from_proto(data_source: DataSourceProto) -> Any:
             file_format=spark_options.file_format,
             event_timestamp_column=data_source.event_timestamp_column,
             created_timestamp_column=data_source.created_timestamp_column,
-            date_partition_column=data_source.date_partition_column,
             description=data_source.description,
             tags=dict(data_source.tags),
             owner=data_source.owner,
@@ -149,7 +157,6 @@ def to_proto(self) -> DataSourceProto:
 
         data_source_proto.event_timestamp_column = self.event_timestamp_column
         data_source_proto.created_timestamp_column = self.created_timestamp_column
-        data_source_proto.date_partition_column = self.date_partition_column
 
         return data_source_proto
 

sdk/python/feast/infra/offline_stores/file_source.py

@@ -66,12 +66,20 @@ def __init__(
             s3_endpoint_override=s3_endpoint_override,
         )
 
+        if date_partition_column:
+            warnings.warn(
+                (
+                    "The argument 'date_partition_column' is not supported for File sources."
+                    "It will be removed in Feast 0.21+"
+                ),
+                DeprecationWarning,
+            )
+
         super().__init__(
-            name if name else path,
-            event_timestamp_column,
-            created_timestamp_column,
-            field_mapping,
-            date_partition_column,
+            name=name if name else path,
+            event_timestamp_column=event_timestamp_column,
+            created_timestamp_column=created_timestamp_column,
+            field_mapping=field_mapping,
             description=description,
             tags=tags,
             owner=owner,
@@ -114,7 +122,6 @@ def from_proto(data_source: DataSourceProto):
             path=data_source.file_options.uri,
             event_timestamp_column=data_source.event_timestamp_column,
             created_timestamp_column=data_source.created_timestamp_column,
-            date_partition_column=data_source.date_partition_column,
             s3_endpoint_override=data_source.file_options.s3_endpoint_override,
             description=data_source.description,
             tags=dict(data_source.tags),
@@ -134,7 +141,6 @@ def to_proto(self) -> DataSourceProto:
 
         data_source_proto.event_timestamp_column = self.event_timestamp_column
         data_source_proto.created_timestamp_column = self.created_timestamp_column
-        data_source_proto.date_partition_column = self.date_partition_column
 
         return data_source_proto
 

sdk/python/feast/infra/offline_stores/redshift_source.py

@@ -21,7 +21,7 @@ def __init__(
         schema: Optional[str] = None,
         created_timestamp_column: Optional[str] = "",
         field_mapping: Optional[Dict[str, str]] = None,
-        date_partition_column: Optional[str] = "",
+        date_partition_column: Optional[str] = None,
         query: Optional[str] = None,
         name: Optional[str] = None,
         description: Optional[str] = "",
@@ -41,7 +41,7 @@ def __init__(
                 row was created, used for deduplicating rows.
             field_mapping (optional): A dictionary mapping of column names in this data
                 source to column names in a feature table or view.
-            date_partition_column (optional): Timestamp column used for partitioning.
+            date_partition_column (deprecated): Timestamp column used for partitioning.
             query (optional): The query to be executed to obtain the features.
             name (optional): Name for the source. Defaults to the table_ref if not specified.
             description (optional): A human-readable description.
@@ -70,13 +70,20 @@ def __init__(
                     ),
                     DeprecationWarning,
                 )
+        if date_partition_column:
+            warnings.warn(
+                (
+                    "The argument 'date_partition_column' is not supported for Redshift sources."
+                    "It will be removed in Feast 0.21+"
+                ),
+                DeprecationWarning,
+            )
 
         super().__init__(
             _name if _name else "",
-            event_timestamp_column,
-            created_timestamp_column,
-            field_mapping,
-            date_partition_column,
+            event_timestamp_column=event_timestamp_column,
+            created_timestamp_column=created_timestamp_column,
+            field_mapping=field_mapping,
             description=description,
             tags=tags,
             owner=owner,
@@ -99,7 +106,6 @@ def from_proto(data_source: DataSourceProto):
             schema=data_source.redshift_options.schema,
             event_timestamp_column=data_source.event_timestamp_column,
             created_timestamp_column=data_source.created_timestamp_column,
-            date_partition_column=data_source.date_partition_column,
             query=data_source.redshift_options.query,
             description=data_source.description,
             tags=dict(data_source.tags),
@@ -169,7 +175,6 @@ def to_proto(self) -> DataSourceProto:
 
         data_source_proto.event_timestamp_column = self.event_timestamp_column
         data_source_proto.created_timestamp_column = self.created_timestamp_column
-        data_source_proto.date_partition_column = self.date_partition_column
 
         return data_source_proto
 

sdk/python/feast/infra/offline_stores/snowflake_source.py

@@ -20,9 +20,9 @@ def __init__(
         table: Optional[str] = None,
         query: Optional[str] = None,
         event_timestamp_column: Optional[str] = "",
+        date_partition_column: Optional[str] = None,
         created_timestamp_column: Optional[str] = "",
         field_mapping: Optional[Dict[str, str]] = None,
-        date_partition_column: Optional[str] = "",
         name: Optional[str] = None,
         description: Optional[str] = "",
         tags: Optional[Dict[str, str]] = None,
@@ -42,7 +42,7 @@ def __init__(
                 row was created, used for deduplicating rows.
             field_mapping (optional): A dictionary mapping of column names in this data
                 source to column names in a feature table or view.
-            date_partition_column (optional): Timestamp column used for partitioning.
+            date_partition_column (deprecated): Timestamp column used for partitioning.
             name (optional): Name for the source. Defaults to the table if not specified.
             description (optional): A human-readable description.
             tags (optional): A dictionary of key-value pairs to store arbitrary metadata.
@@ -72,12 +72,20 @@ def __init__(
                     DeprecationWarning,
                 )
 
+        if date_partition_column:
+            warnings.warn(
+                (
+                    "The argument 'date_partition_column' is not supported for Snowflake sources."
+                    "It will be removed in Feast 0.21+"
+                ),
+                DeprecationWarning,
+            )
+
         super().__init__(
             _name if _name else "",
-            event_timestamp_column,
-            created_timestamp_column,
-            field_mapping,
-            date_partition_column,
+            event_timestamp_column=event_timestamp_column,
+            created_timestamp_column=created_timestamp_column,
+            field_mapping=field_mapping,
             description=description,
             tags=tags,
             owner=owner,
@@ -101,7 +109,6 @@ def from_proto(data_source: DataSourceProto):
             table=data_source.snowflake_options.table,
             event_timestamp_column=data_source.event_timestamp_column,
             created_timestamp_column=data_source.created_timestamp_column,
-            date_partition_column=data_source.date_partition_column,
             query=data_source.snowflake_options.query,
             description=data_source.description,
             tags=dict(data_source.tags),
@@ -170,7 +177,6 @@ def to_proto(self) -> DataSourceProto:
 
         data_source_proto.event_timestamp_column = self.event_timestamp_column
         data_source_proto.created_timestamp_column = self.created_timestamp_column
-        data_source_proto.date_partition_column = self.date_partition_column
 
         return data_source_proto
 

sdk/python/tests/integration/feature_repos/universal/data_sources/bigquery.py

@@ -77,7 +77,6 @@ def create_data_source(
             table_ref=destination_name,
             event_timestamp_column=event_timestamp_column,
             created_timestamp_column=created_timestamp_column,
-            date_partition_column="",
             field_mapping=field_mapping or {"ts_1": "ts"},
         )
 

sdk/python/tests/integration/feature_repos/universal/data_sources/file.py

@@ -48,7 +48,6 @@ def create_data_source(
             path=f"{f.name}",
             event_timestamp_column=event_timestamp_column,
             created_timestamp_column=created_timestamp_column,
-            date_partition_column="",
             field_mapping=field_mapping or {"ts_1": "ts"},
         )
 
@@ -130,7 +129,6 @@ def create_data_source(
             path=f"s3://{self.bucket}/{filename}",
             event_timestamp_column=event_timestamp_column,
             created_timestamp_column=created_timestamp_column,
-            date_partition_column="",
             field_mapping=field_mapping or {"ts_1": "ts"},
             s3_endpoint_override=f"http://{host}:{port}",
         )

sdk/python/tests/integration/feature_repos/universal/data_sources/redshift.py

@@ -63,7 +63,6 @@ def create_data_source(
             table=destination_name,
             event_timestamp_column=event_timestamp_column,
             created_timestamp_column=created_timestamp_column,
-            date_partition_column="",
             field_mapping=field_mapping or {"ts_1": "ts"},
             database=self.offline_store_config.database,
         )

sdk/python/tests/integration/feature_repos/universal/data_sources/snowflake.py

@@ -55,7 +55,6 @@ def create_data_source(
             table=destination_name,
             event_timestamp_column=event_timestamp_column,
             created_timestamp_column=created_timestamp_column,
-            date_partition_column="",
             field_mapping=field_mapping or {"ts_1": "ts"},
         )
 

sdk/python/tests/integration/feature_repos/universal/data_sources/spark_data_source_creator.py

@@ -95,7 +95,6 @@ def create_data_source(
             table=destination_name,
             event_timestamp_column=event_timestamp_column,
             created_timestamp_column=created_timestamp_column,
-            date_partition_column="",
             # maps certain column names to other names
             field_mapping=field_mapping or {"ts_1": "ts"},
         )