REFACTOR-modin-project#6071: Push first and last down to query compil…

…er. (modin-project#64) (modin-project#6125)

Signed-off-by: mvashishtha <mahesh@ponder.io>

modin/core/storage_formats/base/query_compiler.py

@@ -4392,6 +4392,44 @@ def dt_year(self):
 
     # End of DateTime methods
 
+    def first(self, offset: pandas.DateOffset):
+        """
+        Select initial periods of time series data based on a date offset.
+
+        When having a query compiler with dates as index, this function can
+        select the first few rows based on a date offset.
+
+        Parameters
+        ----------
+        offset : pandas.DateOffset
+            The offset length of the data to select.
+
+        Returns
+        -------
+        BaseQueryCompiler
+            New compiler containing the selected data.
+        """
+        return DataFrameDefault.register(pandas.DataFrame.first)(self, offset)
+
+    def last(self, offset: pandas.DateOffset):
+        """
+        Select final periods of time series data based on a date offset.
+
+        For a query compiler with a sorted DatetimeIndex, this function
+        selects the last few rows based on a date offset.
+
+        Parameters
+        ----------
+        offset : pandas.DateOffset
+            The offset length of the data to select.
+
+        Returns
+        -------
+        BaseQueryCompiler
+            New compiler containing the selected data.
+        """
+        return DataFrameDefault.register(pandas.DataFrame.last)(self, offset)
+
     # Resample methods
 
     # FIXME:

modin/pandas/base.py

@@ -10,7 +10,6 @@
 # the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF
 # ANY KIND, either express or implied. See the License for the specific language
 # governing permissions and limitations under the License.
-
 """Implement DataFrame/Series public API as pandas does."""
 from __future__ import annotations
 import numpy as np
@@ -39,6 +38,7 @@
 )
 from pandas._libs.lib import NoDefault
 from pandas._libs.lib import no_default
+from pandas._libs.tslibs import to_offset
 from pandas._typing import (
     IndexKeyFunc,
     StorageOptions,
@@ -1628,7 +1628,9 @@ def first(self, offset):  # noqa: PR01, RT01, D200
         """
         Select initial periods of time series data based on a date offset.
         """
-        return self.loc[pandas.Series(index=self.index).first(offset).index]
+        return self._create_or_update_from_compiler(
+            self._query_compiler.first(offset=to_offset(offset))
+        )
 
     def first_valid_index(self):  # noqa: RT01, D200
         """
@@ -1809,7 +1811,9 @@ def last(self, offset):  # noqa: PR01, RT01, D200
         """
         Select final periods of time series data based on a date offset.
         """
-        return self.loc[pandas.Series(index=self.index).last(offset).index]
+        return self._create_or_update_from_compiler(
+            self._query_compiler.last(offset=to_offset(offset))
+        )
 
     def last_valid_index(self):  # noqa: RT01, D200
         """