FIX-#7638: Suppress default to pandas warnings on native pandas backend

modin/core/storage_formats/base/query_compiler.py

@@ -206,7 +206,8 @@ class BaseQueryCompiler(
     _shape_hint: Optional[str]
     _should_warn_on_default_to_pandas: bool = True
 
-    def _maybe_warn_on_default(self, *, message: str = "", reason: str = "") -> None:
+    @classmethod
+    def _maybe_warn_on_default(cls, *, message: str = "", reason: str = "") -> None:
         """
         If this class is configured to warn on default to pandas, warn.
 
@@ -217,7 +218,7 @@ def _maybe_warn_on_default(self, *, message: str = "", reason: str = "") -> None
         reason : str, default: ""
             Reason for default.
         """
-        if self._should_warn_on_default_to_pandas:
+        if cls._should_warn_on_default_to_pandas:
             ErrorMessage.default_to_pandas(message=message, reason=reason)
 
     @disable_logging

modin/pandas/base.py

@@ -561,8 +561,8 @@ def _default_to_pandas(self, op, *args, reason: str = None, **kwargs):
             Result of operation.
         """
         empty_self_str = "" if not self.empty else " for empty DataFrame"
-        ErrorMessage.default_to_pandas(
-            "`{}.{}`{}".format(
+        self._query_compiler._maybe_warn_on_default(
+            message="`{}.{}`{}".format(
                 type(self).__name__,
                 op if isinstance(op, str) else op.__name__,
                 empty_self_str,

modin/pandas/general.py

@@ -28,10 +28,9 @@
 from modin.core.storage_formats.pandas.query_compiler_caster import (
     wrap_free_function_in_argument_caster,
 )
-from modin.error_message import ErrorMessage
 from modin.logging import enable_logging
 from modin.pandas.io import to_pandas
-from modin.utils import _inherit_docstrings
+from modin.utils import _inherit_docstrings, _maybe_warn_on_default
 
 from .base import BasePandasDataset
 from .dataframe import DataFrame
@@ -193,7 +192,7 @@ def merge_asof(
         raise ValueError(
             "can not merge DataFrame with instance of type {}".format(type(right))
         )
-    ErrorMessage.default_to_pandas("`merge_asof`")
+    left._query_compiler._maybe_warn_on_default(message="`merge_asof`")
 
     # As of Pandas 1.2 these should raise an error; before that it did
     # something likely random:
@@ -345,7 +344,7 @@ def cut(
     if isinstance(x, DataFrame):
         raise ValueError("Input array must be 1 dimensional")
     if not isinstance(x, Series):
-        ErrorMessage.default_to_pandas(
+        _maybe_warn_on_default(
             reason=f"pd.cut is not supported on objects of type {type(x)}"
         )
         import pandas
@@ -656,7 +655,7 @@ def get_dummies(
             + "github.com/modin-project/modin."
         )
     if not isinstance(data, DataFrame):
-        ErrorMessage.default_to_pandas("`get_dummies` on non-DataFrame")
+        _maybe_warn_on_default("`get_dummies` on non-DataFrame")
         if isinstance(data, Series):
             data = data._to_pandas()
         return DataFrame(
@@ -726,7 +725,7 @@ def crosstab(
     """
     Compute a simple cross tabulation of two (or more) factors.
     """
-    ErrorMessage.default_to_pandas("`crosstab`")
+    _maybe_warn_on_default("`crosstab`")
     pandas_crosstab = pandas.crosstab(
         index,
         columns,
@@ -769,7 +768,7 @@ def lreshape(data: DataFrame, groups, dropna=True) -> DataFrame:
     """
     if not isinstance(data, DataFrame):
         raise ValueError("can not lreshape with instance of type {}".format(type(data)))
-    ErrorMessage.default_to_pandas("`lreshape`")
+    data._query_compiler._maybe_warn_on_default(message="`lreshape`")
     return DataFrame(pandas.lreshape(to_pandas(data), groups, dropna=dropna))
 
 

modin/pandas/io.py

@@ -71,13 +71,13 @@
 from modin.core.storage_formats.pandas.query_compiler_caster import (
     wrap_free_function_in_argument_caster,
 )
-from modin.error_message import ErrorMessage
 from modin.logging import ClassLogger, enable_logging
 from modin.utils import (
     SupportsPrivateToNumPy,
     SupportsPublicToNumPy,
     SupportsPublicToPandas,
     _inherit_docstrings,
+    _maybe_warn_on_default,
     classproperty,
     expanduser_path_arg,
 )
@@ -156,7 +156,7 @@ def read_xml(
     storage_options: StorageOptions = None,
     dtype_backend: Union[DtypeBackend, NoDefault] = no_default,
 ) -> DataFrame:
-    ErrorMessage.default_to_pandas("read_xml")
+    _maybe_warn_on_default("read_xml")
     _, _, _, kwargs = inspect.getargvalues(inspect.currentframe())
     return ModinObjects.DataFrame(pandas.read_xml(**kwargs))
 
@@ -658,7 +658,7 @@ def read_sql(
     from modin.core.execution.dispatching.factories.dispatcher import FactoryDispatcher
 
     if kwargs.get("chunksize") is not None:
-        ErrorMessage.default_to_pandas("Parameters provided [chunksize]")
+        _maybe_warn_on_default("Parameters provided [chunksize]")
         df_gen = pandas.read_sql(**kwargs)
         return (
             ModinObjects.DataFrame(query_compiler=FactoryDispatcher.from_pandas(df))
@@ -818,7 +818,7 @@ def json_normalize(
     """
     Normalize semi-structured JSON data into a flat table.
     """
-    ErrorMessage.default_to_pandas("json_normalize")
+    _maybe_warn_on_default("json_normalize")
     return ModinObjects.DataFrame(
         pandas.json_normalize(
             data, record_path, meta, meta_prefix, record_prefix, errors, sep, max_level
@@ -840,7 +840,7 @@ def read_orc(
     """
     Load an ORC object from the file path, returning a DataFrame.
     """
-    ErrorMessage.default_to_pandas("read_orc")
+    _maybe_warn_on_default("read_orc")
     return ModinObjects.DataFrame(
         pandas.read_orc(
             path,
@@ -886,7 +886,7 @@ def return_handler(*args, **kwargs):
                     # We don't want to constantly be giving this error message for
                     # internal methods.
                     if item[0] != "_":
-                        ErrorMessage.default_to_pandas("`{}`".format(item))
+                        _maybe_warn_on_default("`{}`".format(item))
                     args = [
                         (
                             to_pandas(arg)
@@ -952,7 +952,7 @@ def return_handler(*args, **kwargs):
                     # We don't want to constantly be giving this error message for
                     # internal methods.
                     if item[0] != "_":
-                        ErrorMessage.default_to_pandas("`{}`".format(item))
+                        _maybe_warn_on_default("`{}`".format(item))
                     args = [
                         (
                             to_pandas(arg)

modin/tests/experimental/test_io_exp.py

@@ -29,7 +29,7 @@
     time_parsing_csv_path,
 )
 from modin.tests.test_utils import (
-    warns_that_defaulting_to_pandas,
+    current_execution_is_native,
     warns_that_defaulting_to_pandas_if,
 )
 from modin.utils import try_cast_to_pandas
@@ -129,7 +129,7 @@ def test_read_csv_empty_frame(self):
 
     def test_read_csv_without_glob(self):
         with pytest.raises(FileNotFoundError):
-            with warns_that_defaulting_to_pandas():
+            with warns_that_defaulting_to_pandas_if(not current_execution_is_native()):
                 pd.read_csv_glob(
                     "s3://dask-data/nyc-taxi/2015/yellow_tripdata_2015-",
                     storage_options={"anon": True},

modin/tests/interchange/dataframe_protocol/pandas/test_protocol.py

@@ -20,7 +20,6 @@
 from modin.tests.pandas.utils import df_equals, test_data
 from modin.tests.test_utils import (
     df_or_series_using_native_execution,
-    warns_that_defaulting_to_pandas,
     warns_that_defaulting_to_pandas_if,
 )
 
@@ -66,7 +65,9 @@ def test_categorical_from_dataframe():
 
 def test_from_dataframe_with_empty_dataframe():
     modin_df = pd.DataFrame({"foo_col": pd.Series([], dtype="int64")})
-    with warns_that_defaulting_to_pandas():
+    with warns_that_defaulting_to_pandas_if(
+        not df_or_series_using_native_execution(modin_df)
+    ):
         eval_df_protocol(modin_df)
 
 

modin/tests/pandas/dataframe/test_iter.py

@@ -35,7 +35,6 @@
 )
 from modin.tests.test_utils import (
     current_execution_is_native,
-    warns_that_defaulting_to_pandas,
     warns_that_defaulting_to_pandas_if,
 )
 
@@ -147,7 +146,7 @@ def test_display_options_for___repr__(max_rows_columns, expand_frame_repr, frame
 def test___finalize__():
     data = test_data_values[0]
     # NOTE: __finalize__() defaults to pandas at the API layer.
-    with warns_that_defaulting_to_pandas():
+    with warns_that_defaulting_to_pandas_if(not current_execution_is_native()):
         pd.DataFrame(data).__finalize__(None)
 
 

modin/tests/pandas/dataframe/test_udf.py

@@ -43,7 +43,10 @@
     udf_func_keys,
     udf_func_values,
 )
-from modin.tests.test_utils import warns_that_defaulting_to_pandas
+from modin.tests.test_utils import (
+    current_execution_is_native,
+    warns_that_defaulting_to_pandas_if,
+)
 from modin.utils import get_current_execution
 
 NPartitions.put(4)
@@ -126,10 +129,10 @@ def test_aggregate_alias():
 def test_aggregate_error_checking():
     modin_df = pd.DataFrame(test_data["float_nan_data"])
 
-    with warns_that_defaulting_to_pandas():
+    with warns_that_defaulting_to_pandas_if(not current_execution_is_native()):
         modin_df.aggregate({modin_df.columns[0]: "sum", modin_df.columns[1]: "mean"})
 
-    with warns_that_defaulting_to_pandas():
+    with warns_that_defaulting_to_pandas_if(not current_execution_is_native()):
         modin_df.aggregate("arcsin")
 
 

modin/tests/pandas/extensions/test_groupby_extensions.py

@@ -24,7 +24,10 @@
 )
 from modin.pandas.groupby import DataFrameGroupBy, SeriesGroupBy
 from modin.tests.pandas.utils import default_to_pandas_ignore_string, df_equals
-from modin.tests.test_utils import warns_that_defaulting_to_pandas
+from modin.tests.test_utils import (
+    current_execution_is_native,
+    warns_that_defaulting_to_pandas_if,
+)
 
 
 @pytest.mark.parametrize(
@@ -150,10 +153,7 @@ def ngroups(self):
         # Check that the accessor doesn't work on the Python_Test backend.
         python_test_df = pandas_df.move_to("Python_Test")
         groupby = get_groupby(python_test_df)
-        # groupby.ngroups defaults to pandas at the API layer,
-        # where it warns that it's doing so, even for dataframes using the
-        # Pandas backend.
-        with warns_that_defaulting_to_pandas():
+        with warns_that_defaulting_to_pandas_if(not current_execution_is_native()):
             assert groupby.ngroups == 3
 
     def test_add_ngroups_setter_and_deleter_for_one_backend(
@@ -179,7 +179,7 @@ def _set_ngroups(self, value):
 
         python_test_groupby = get_groupby(python_test_df)
 
-        with warns_that_defaulting_to_pandas():
+        with warns_that_defaulting_to_pandas_if(not current_execution_is_native()):
             assert python_test_groupby.ngroups == 3
 
         with pytest.raises(AttributeError):

modin/tests/pandas/native_df_interoperability/test_default_to_pandas_without_warnings.py

@@ -0,0 +1,78 @@
+# Licensed to Modin Development Team under one or more contributor license agreements.
+# See the NOTICE file distributed with this work for additional information regarding
+# copyright ownership.  The Modin Development Team licenses this file to you under the
+# Apache License, Version 2.0 (the "License"); you may not use this file except in
+# compliance with the License.  You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software distributed under
+# 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.
+
+# While other modin backends raise a warning when defaulting to pandas, it does not make sense to
+# do so when we're running on the native pandas backend already. These tests ensure such warnings
+# are not raised with the pandas backend.
+
+import numpy as np
+import pandas
+import pytest
+
+import modin.pandas as pd
+from modin.config import Backend
+from modin.tests.pandas.utils import df_equals
+
+pytestmark = [
+    pytest.mark.skipif(
+        Backend.get() != "Pandas",
+        reason="warnings only suppressed on native pandas backend",
+        allow_module_level=True,
+    ),
+    # Error if a default to pandas warning is detected.
+    pytest.mark.filterwarnings("error:is not supported by NativeOnNative:UserWarning"),
+]
+
+
+def test_crosstab_no_warning():
+    # Example from pandas docs
+    # https://pandas.pydata.org/docs/reference/api/pandas.crosstab.html
+    a = np.array(
+        ["foo", "foo", "foo", "foo", "bar", "bar", "bar", "bar", "foo", "foo", "foo"],
+        dtype=object,
+    )
+    b = np.array(
+        ["one", "one", "one", "two", "one", "one", "one", "two", "two", "two", "one"],
+        dtype=object,
+    )
+    c = np.array(
+        [
+            "dull",
+            "dull",
+            "shiny",
+            "dull",
+            "dull",
+            "shiny",
+            "shiny",
+            "dull",
+            "shiny",
+            "shiny",
+            "shiny",
+        ],
+        dtype=object,
+    )
+    df_equals(
+        pd.crosstab(a, [b, c], rownames=["a"], colnames=["b", "c"]),
+        pandas.crosstab(a, [b, c], rownames=["a"], colnames=["b", "c"]),
+    )
+
+
+def test_json_normalize_no_warning():
+    # Example from pandas docs
+    # https://pandas.pydata.org/docs/reference/api/pandas.json_normalize.html
+    data = [
+        {"id": 1, "name": {"first": "Coleen", "last": "Volk"}},
+        {"name": {"given": "Mark", "family": "Regner"}},
+        {"id": 2, "name": "Faye Raker"},
+    ]
+    df_equals(pd.json_normalize(data), pandas.json_normalize(data))