Merge pull request #329 from lvgig/feature/tests_dtseries

refactored SeriesDtTransformer tests

Author: Chip2916

CHANGELOG.rst

@@ -43,6 +43,7 @@ Changed
 - fixed a bug in CappingTransformer which was preventing use of .get_params method `#311 <https://github.com/lvgig/tubular/issues/311>`_
 - Setup requirements for narwhals, remove python3.8 from our build pipelines as incompatible with polars
 - Refactored ToDatetimeTransformer tests in new format `#300 <https://github.com/lvgig/tubular/issues/300>`_
+- Refactors tests for SeriesDtMethodTransformer in new format. Changed column arg to columns to fit generic format. `#299 <https://github.com/lvgig/tubular/issues/299>_`
 - Refactored OrdinalEncoderTransformer tests in new format `#330 <https://github.com/lvgig/tubular/issues/330>`_
 
 

tests/conftest.py

@@ -250,9 +250,9 @@ def minimal_attribute_dict():
             "columns": ["a", "b"],
         },
         "SeriesDtMethodTransformer": {
-            "new_column_name": "a",
+            "new_column_name": "new_column",
             "pd_method_name": "month",
-            "column": "b",
+            "columns": "b",
         },
         "SeriesStrMethodTransformer": {
             "columns": ["b"],

tests/dates/test_SeriesDtMethodTransformer.py

@@ -1,40 +1,56 @@
+import re
+
 import numpy as np
 import pytest
 import test_aide as ta
 
 import tests.test_data as d
+from tests.base_tests import (
+    ColumnStrListInitTests,
+    DropOriginalInitMixinTests,
+    DropOriginalTransformMixinTests,
+    GenericTransformTests,
+    NewColumnNameInitMixintests,
+    OtherBaseBehaviourTests,
+)
+from tests.dates.test_BaseDatetimeTransformer import (
+    DatetimeMixinTransformTests,
+)
 from tubular.dates import SeriesDtMethodTransformer
 
 
-class TestInit:
+class TestInit(
+    ColumnStrListInitTests,
+    DropOriginalInitMixinTests,
+    NewColumnNameInitMixintests,
+):
     """Tests for SeriesDtMethodTransformer.init()."""
 
+    @classmethod
+    def setup_class(cls):
+        cls.transformer_name = "SeriesDtMethodTransformer"
+
     def test_invalid_input_type_errors(self):
         """Test that an exceptions are raised for invalid input types."""
+        bad_columns = ["b", "c"]
         with pytest.raises(
-            TypeError,
-            match=r"SeriesDtMethodTransformer: column should be a str but got \<class 'list'\>",
+            ValueError,
+            match=rf"SeriesDtMethodTransformer: column should be a str or list of len 1, got {re.escape(str(bad_columns))}",
         ):
             SeriesDtMethodTransformer(
                 new_column_name="a",
                 pd_method_name=1,
-                column=["b", "c"],
+                columns=bad_columns,
             )
 
         with pytest.raises(
             TypeError,
             match=r"SeriesDtMethodTransformer: unexpected type \(\<class 'int'\>\) for pd_method_name, expecting str",
-        ):
-            SeriesDtMethodTransformer(new_column_name="a", pd_method_name=1, column="b")
-
-        with pytest.raises(
-            TypeError,
-            match=r"SeriesDtMethodTransformer: new_column_name should be str",
         ):
             SeriesDtMethodTransformer(
-                new_column_name=1.0,
-                pd_method_name="year",
-                column="b",
+                new_column_name="a",
+                pd_method_name=1,
+                columns="b",
             )
 
         with pytest.raises(
@@ -44,7 +60,7 @@ def test_invalid_input_type_errors(self):
             SeriesDtMethodTransformer(
                 new_column_name="a",
                 pd_method_name="year",
-                column="b",
+                columns="b",
                 pd_method_kwargs=1,
             )
 
@@ -55,7 +71,7 @@ def test_invalid_input_type_errors(self):
             SeriesDtMethodTransformer(
                 new_column_name="a",
                 pd_method_name="year",
-                column="b",
+                columns="b",
                 pd_method_kwargs={"a": 1, 2: "b"},
             )
 
@@ -68,52 +84,21 @@ def test_exception_raised_non_pandas_method_passed(self):
             SeriesDtMethodTransformer(
                 new_column_name="a",
                 pd_method_name="b",
-                column="b",
+                columns="b",
             )
 
-    def test_attributes_set(self):
-        """Test that the values passed for new_column_name, pd_method_name are saved to attributes on the object."""
-        x = SeriesDtMethodTransformer(
-            new_column_name="a",
-            pd_method_name="year",
-            column="b",
-            pd_method_kwargs={"d": 1},
-        )
-
-        ta.classes.test_object_attributes(
-            obj=x,
-            expected_attributes={
-                "column": "b",
-                "new_column_name": "a",
-                "pd_method_name": "year",
-                "pd_method_kwargs": {"d": 1},
-            },
-            msg="Attributes for SeriesDtMethodTransformer set in init",
-        )
 
-    @pytest.mark.parametrize(
-        ("pd_method_name", "callable_attr"),
-        [("year", False), ("to_period", True)],
-    )
-    def test_callable_attribute_set(self, pd_method_name, callable_attr):
-        """Test the _callable attribute is set to True if pd.Series.dt.pd_method_name is callable."""
-        x = SeriesDtMethodTransformer(
-            new_column_name="a",
-            pd_method_name=pd_method_name,
-            column="b",
-            pd_method_kwargs={"d": 1},
-        )
-
-        ta.classes.test_object_attributes(
-            obj=x,
-            expected_attributes={"_callable": callable_attr},
-            msg="_callable attribute for SeriesDtMethodTransformer set in init",
-        )
-
-
-class TestTransform:
+class TestTransform(
+    DatetimeMixinTransformTests,
+    DropOriginalTransformMixinTests,
+    GenericTransformTests,
+):
     """Tests for SeriesDtMethodTransformer.transform()."""
 
+    @classmethod
+    def setup_class(cls):
+        cls.transformer_name = "SeriesDtMethodTransformer"
+
     def expected_df_1():
         """Expected output of test_expected_output_no_overwrite."""
         df = d.create_datediff_test_df()
@@ -144,31 +129,6 @@ def expected_df_3():
 
         return df
 
-    @pytest.mark.parametrize(
-        ("bad_column", "bad_type"),
-        [
-            ("numeric_col", "int64"),
-            ("string_col", "object"),
-            ("bool_col", "bool"),
-            ("empty_col", "object"),
-            ("date_col", "date"),
-        ],
-    )
-    def test_input_data_check_column_errors(self, bad_column, bad_type):
-        """Check that errors are raised on a variety of different non datatypes"""
-        x = SeriesDtMethodTransformer(
-            new_column_name="a2",
-            pd_method_name="year",
-            column=bad_column,
-        )
-
-        df = d.create_date_diff_incorrect_dtypes()
-
-        msg = rf"{x.classname()}: {x.columns[0]} type should be in \['datetime64'\] but got {bad_type}"
-
-        with pytest.raises(TypeError, match=msg):
-            x.transform(df)
-
     @pytest.mark.parametrize(
         ("df", "expected"),
         ta.pandas.adjusted_dataframe_params(
@@ -181,7 +141,7 @@ def test_expected_output_no_overwrite(self, df, expected):
         x = SeriesDtMethodTransformer(
             new_column_name="a_year",
             pd_method_name="year",
-            column="a",
+            columns="a",
             pd_method_kwargs=None,
         )
 
@@ -205,7 +165,7 @@ def test_expected_output_overwrite(self, df, expected):
         x = SeriesDtMethodTransformer(
             new_column_name="a",
             pd_method_name="year",
-            column="a",
+            columns="a",
             pd_method_kwargs=None,
         )
 
@@ -229,7 +189,7 @@ def test_expected_output_callable(self, df, expected):
         x = SeriesDtMethodTransformer(
             new_column_name="b_new",
             pd_method_name="to_period",
-            column="b",
+            columns="b",
             pd_method_kwargs={"freq": "M"},
         )
 
@@ -241,35 +201,14 @@ def test_expected_output_callable(self, df, expected):
             msg_tag="Unexpected values in SeriesDtMethodTransformer.transform with to_period",
         )
 
-    def test_attributes_unchanged_by_transform(self):
-        """Test that attributes set in init are unchanged by the transform method."""
-        df = d.create_datediff_test_df()
 
-        x = SeriesDtMethodTransformer(
-            new_column_name="b_new",
-            pd_method_name="to_period",
-            column="b",
-            pd_method_kwargs={"freq": "M"},
-        )
+class TestOtherBaseBehaviour(OtherBaseBehaviourTests):
+    """
+    Class to run tests for BaseTransformerBehaviour outside the three standard methods.
 
-        x2 = SeriesDtMethodTransformer(
-            new_column_name="b_new",
-            pd_method_name="to_period",
-            column="b",
-            pd_method_kwargs={"freq": "M"},
-        )
+    May need to overwite specific tests in this class if the tested transformer modifies this behaviour.
+    """
 
-        x.transform(df)
-
-        assert (
-            x.new_column_name == x2.new_column_name
-        ), "new_column_name changed by SeriesDtMethodTransformer.transform"
-        assert (
-            x.pd_method_name == x2.pd_method_name
-        ), "pd_method_name changed by SeriesDtMethodTransformer.transform"
-        assert (
-            x.columns == x2.columns
-        ), "columns changed by SeriesDtMethodTransformer.transform"
-        assert (
-            x.pd_method_kwargs == x2.pd_method_kwargs
-        ), "pd_method_kwargs changed by SeriesDtMethodTransformer.transform"
+    @classmethod
+    def setup_class(cls):
+        cls.transformer_name = "SeriesDtMethodTransformer"

tubular/dates.py

@@ -591,22 +591,24 @@ def __init__(
         self,
         new_column_name: str,
         pd_method_name: str,
-        column: str,
+        columns: list[str],
         pd_method_kwargs: dict[str, object] | None = None,
         drop_original: bool = False,
         **kwargs: dict[str, bool],
     ) -> None:
-        if type(column) is not str:
-            msg = f"{self.classname()}: column should be a str but got {type(column)}"
-            raise TypeError(msg)
-
         super().__init__(
-            columns=[column],
+            columns=columns,
             new_column_name=new_column_name,
             drop_original=drop_original,
             **kwargs,
         )
 
+        if len(self.columns) > 1:
+            msg = rf"{self.classname()}: column should be a str or list of len 1, got {self.columns}"
+            raise ValueError(
+                msg,
+            )
+
         if type(pd_method_name) is not str:
             msg = f"{self.classname()}: unexpected type ({type(pd_method_name)}) for pd_method_name, expecting str"
             raise TypeError(msg)
@@ -644,7 +646,7 @@ def __init__(
 
         # This attribute is not for use in any method, use 'columns' instead.
         # Here only as a fix to allow string representation of transformer.
-        self.column = column
+        self.column = self.columns[0]
 
     def transform(self, X: pd.DataFrame) -> pd.DataFrame:
         """Transform specific column on input pandas.DataFrame (X) using the given pandas.Series.dt method and