Make public method to get subset schema (#1218)

* apply changes

* update release notes

Author: jeff-hernandez

docs/source/release_notes.rst

@@ -8,11 +8,12 @@ Future Release
         * Add type inference for natural language (:pr:`1210`) 
     * Fixes
     * Changes
+        * Make public method ``get_subset_schema`` (:pr:`1218`)
     * Documentation Changes
     * Testing Changes
 
    Thanks to the following people for contributing to this release:
-   :user:`tuethan1999`
+   :user:`jeff-hernandez`, :user:`tuethan1999`
 
 v0.10.0 Nov 30, 2020
 ====================

woodwork/indexers.py

@@ -53,7 +53,7 @@ def _process_selection(selection, original_data):
     elif _is_dataframe(selection):
         # Selecting a new DataFrame from an existing DataFrame
         schema = original_data.ww.schema
-        new_schema = schema._get_subset_schema(list(selection.columns))
+        new_schema = schema.get_subset_schema(list(selection.columns))
         selection.ww.init_with_full_schema(schema=new_schema, validate=False)
     # Selecting a single value or return selection from above
     return selection

woodwork/table_accessor.py

@@ -577,7 +577,7 @@ def select(self, include=None, exclude=None, return_schema=False):
         cols_to_include = self._schema._filter_cols(include, exclude)
 
         if return_schema:
-            return self._schema._get_subset_schema(cols_to_include)
+            return self._schema.get_subset_schema(cols_to_include)
         return self._get_subset_df_with_schema(cols_to_include)
 
     @_check_table_schema
@@ -776,7 +776,7 @@ def _get_subset_df_with_schema(self, cols_to_include, inplace=False):
 
         assert all([col_name in self._schema.columns for col_name in cols_to_include])
 
-        new_schema = self._schema._get_subset_schema(cols_to_include)
+        new_schema = self._schema.get_subset_schema(cols_to_include)
         if inplace:
             cols_to_drop = [
                 col_name

woodwork/table_schema.py

@@ -552,7 +552,12 @@ def _filter_cols(self, include=None, exclude=None, col_names=False):
 
         return cols_to_return
 
-    def _get_subset_schema(self, subset_cols):
+    @property
+    def _get_subset_schema(self):
+        # removing this property might cause breaking changes in EvalML
+        return self.get_subset_schema
+
+    def get_subset_schema(self, subset_cols):
         """Creates a new TableSchema with specified columns, retaining typing information.
 
         Args:

woodwork/tests/accessor/test_table_accessor.py

@@ -2014,7 +2014,7 @@ def test_pop(sample_df):
     assert "age" not in schema_df.ww.logical_types.keys()
     assert "age" not in schema_df.ww.semantic_tags.keys()
 
-    assert schema_df.ww.schema == original_schema._get_subset_schema(
+    assert schema_df.ww.schema == original_schema.get_subset_schema(
         list(schema_df.columns)
     )
 

woodwork/tests/schema/test_table_schema.py

@@ -520,7 +520,7 @@ def test_get_subset_schema_all_params(
     assert set(possible_schema_params) == set(kwargs.keys())
 
     schema = TableSchema(**kwargs)
-    copy_schema = schema._get_subset_schema(sample_column_names)
+    copy_schema = schema.get_subset_schema(sample_column_names)
 
     assert schema == copy_schema
     assert schema is not copy_schema
@@ -1158,7 +1158,7 @@ def test_set_index_twice(sample_column_names, sample_inferred_logical_types):
         index="id",
         time_index="signup_date",
     )
-    original_schema = schema._get_subset_schema(list(schema.columns.keys()))
+    original_schema = schema.get_subset_schema(list(schema.columns.keys()))
 
     schema.set_index("id")
     assert schema.index == "id"
@@ -1210,7 +1210,7 @@ def test_schema_rename(sample_column_names, sample_inferred_logical_types):
         column_descriptions={"id": id_description},
         column_origins={"id": id_origin},
     )
-    original_schema = schema._get_subset_schema(list(schema.columns.keys()))
+    original_schema = schema.get_subset_schema(list(schema.columns.keys()))
 
     renamed_schema = schema.rename({"age": "birthday"})