Update to mypy 0.971

.pre-commit-config.yaml

@@ -26,10 +26,15 @@ repos:
                 files: python/.*\.(py|pyx|pxd)$
                 types: [file]
       - repo: https://github.com/pre-commit/mirrors-mypy
-        rev: 'v0.782'
+        rev: 'v0.971'
         hooks:
               - id: mypy
-                args: ["--config-file=setup.cfg", "python/cudf/cudf", "python/dask_cudf/dask_cudf", "python/custreamz/custreamz", "python/cudf_kafka/cudf_kafka"]
+                additional_dependencies: [types-cachetools==5.2.1]
+                args: ["--config-file=setup.cfg",
+                       "python/cudf/cudf",
+                       "python/custreamz/custreamz",
+                       "python/cudf_kafka/cudf_kafka",
+                       "python/dask_cudf/dask_cudf"]
                 pass_filenames: false
       - repo: https://github.com/PyCQA/pydocstyle
         rev: 6.1.1

conda/environments/cudf_dev_cuda11.5.yml

@@ -43,7 +43,8 @@ dependencies:
   - flake8=3.8.3
   - black=22.3.0
   - isort=5.10.1
-  - mypy=0.782
+  - mypy=0.971
+  - types-cachetools==5.2.1
   - doxygen=1.8.20
   - pydocstyle=6.1.1
   - typing_extensions

python/cudf/cudf/core/column/string.py

@@ -34,6 +34,7 @@
 )
 from cudf.core.buffer import DeviceBufferLike
 from cudf.core.column import column, datetime
+from cudf.core.column.column import ColumnBase
 from cudf.core.column.methods import ColumnMethods
 from cudf.utils.docutils import copy_docstring
 from cudf.utils.dtypes import can_convert_to_column
@@ -3650,7 +3651,12 @@ def isempty(self) -> SeriesOrIndex:
         4    False
         dtype: bool
         """
-        return self._return_or_inplace((self._column == "").fillna(False))
+        return self._return_or_inplace(
+            # mypy can't deduce that the return value of
+            # StringColumn.__eq__ is ColumnBase because the binops are
+            # dynamically added by a mixin class
+            cast(ColumnBase, self._column == "").fillna(False)
+        )
 
     def isspace(self) -> SeriesOrIndex:
         r"""

python/cudf/cudf/core/dataframe.py

@@ -21,7 +21,6 @@
     Optional,
     Set,
     Tuple,
-    Type,
     TypeVar,
     Union,
 )
@@ -104,6 +103,13 @@
     _external_only_api,
 )
 
+if sys.version_info >= (3, 10):
+    # Introduced in Python 3.10
+    from types import NotImplementedType
+else:
+    NotImplementedType = Any
+
+
 T = TypeVar("T", bound="DataFrame")
 
 
@@ -1934,7 +1940,7 @@ def _make_operands_and_index_for_binop(
     ) -> Tuple[
         Union[
             Dict[Optional[str], Tuple[ColumnBase, Any, bool, Any]],
-            Type[NotImplemented],
+            NotImplementedType,
         ],
         Optional[BaseIndex],
     ]:

python/cudf/cudf/core/index.py

@@ -231,7 +231,7 @@ def step(self):
     def _num_rows(self):
         return len(self)
 
-    @cached_property
+    @cached_property  # type: ignore
     @_cudf_nvtx_annotate
     def _values(self):
         if len(self) > 0:

python/cudf/cudf/core/indexed_frame.py

@@ -5,6 +5,7 @@
 
 import numbers
 import operator
+import sys
 import textwrap
 import warnings
 from collections import Counter, abc
@@ -55,6 +56,13 @@
 from cudf.utils import docutils
 from cudf.utils.utils import _cudf_nvtx_annotate
 
+if sys.version_info >= (3, 10):
+    # Introduced in Python 3.10
+    from types import NotImplementedType
+else:
+    NotImplementedType = Any
+
+
 doc_reset_index_template = """
         Reset the index of the {klass}, or a level of it.
 
@@ -2991,7 +2999,7 @@ def _make_operands_and_index_for_binop(
     ) -> Tuple[
         Union[
             Dict[Optional[str], Tuple[ColumnBase, Any, bool, Any]],
-            Type[NotImplemented],
+            NotImplementedType,
         ],
         Optional[cudf.BaseIndex],
     ]:

python/cudf/cudf/core/multiindex.py

@@ -1458,7 +1458,7 @@ def from_pandas(cls, multiindex, nan_as_null=None):
         )
         return cls.from_frame(df, names=multiindex.names)
 
-    @cached_property
+    @cached_property  # type: ignore
     @_cudf_nvtx_annotate
     def is_unique(self):
         return len(self) == len(self.unique())

python/cudf/cudf/core/series.py

@@ -5,10 +5,11 @@
 import functools
 import inspect
 import pickle
+import sys
 import textwrap
 from collections import abc
 from shutil import get_terminal_size
-from typing import Any, Dict, MutableMapping, Optional, Set, Tuple, Type, Union
+from typing import Any, Dict, MutableMapping, Optional, Set, Tuple, Union
 
 import cupy
 import numpy as np
@@ -74,6 +75,12 @@
 )
 from cudf.utils.utils import _cudf_nvtx_annotate
 
+if sys.version_info >= (3, 10):
+    # Introduced in Python 3.10
+    from types import NotImplementedType
+else:
+    NotImplementedType = Any
+
 
 def _format_percentile_names(percentiles):
     return [f"{int(x * 100)}%" for x in percentiles]
@@ -1289,7 +1296,7 @@ def _make_operands_and_index_for_binop(
     ) -> Tuple[
         Union[
             Dict[Optional[str], Tuple[ColumnBase, Any, bool, Any]],
-            Type[NotImplemented],
+            NotImplementedType,
         ],
         Optional[BaseIndex],
     ]:

python/cudf/cudf/core/single_column_frame.py

@@ -3,8 +3,9 @@
 
 from __future__ import annotations
 
+import sys
 import warnings
-from typing import Any, Dict, Optional, Tuple, Type, TypeVar, Union
+from typing import Any, Dict, Optional, Tuple, TypeVar, Union
 
 import cupy
 import numpy as np
@@ -21,6 +22,13 @@
 from cudf.core.frame import Frame
 from cudf.utils.utils import NotIterable, _cudf_nvtx_annotate
 
+if sys.version_info >= (3, 10):
+    # Introduced in Python 3.10
+    from types import NotImplementedType
+else:
+    NotImplementedType = Any
+
+
 T = TypeVar("T", bound="Frame")
 
 
@@ -302,7 +310,7 @@ def _make_operands_for_binop(
         **kwargs,
     ) -> Union[
         Dict[Optional[str], Tuple[ColumnBase, Any, bool, Any]],
-        Type[NotImplemented],
+        NotImplementedType,
     ]:
         """Generate the dictionary of operands used for a binary operation.
 

setup.cfg

@@ -37,33 +37,15 @@ select =
 
 [mypy]
 ignore_missing_imports = True
-
-[mypy-cudf._lib.*]
-ignore_errors = True
-
-[mypy-cudf._version]
-ignore_errors = True
-
-[mypy-cudf.utils.metadata.orc_column_statistics_pb2]
-ignore_errors = True
-
-[mypy-cudf.tests.*]
-ignore_errors = True
-
-[mypy-dask_cudf._version]
-ignore_errors = True
-
-[mypy-dask_cudf.tests.*]
-ignore_errors = True
-
-[mypy-custreamz._version]
-ignore_errors = True
-
-[mypy-custreamz.tests.*]
-ignore_errors = True
-
-[mypy-cudf_kafka._version]
-ignore_errors = True
-
-[mypy-cudf_kafka.tests.*]
-ignore_errors = True
+# If we don't specify this, then mypy will check excluded files if
+# they are imported by an checked file.
+follow_imports = skip
+exclude = (?x)(
+  (cudf|custreamz|cudf_kafka|dask_cudf)/_version\.py
+  | cudf/_lib/
+  | cudf/cudf/benchmarks/
+  | cudf/cudf/tests/
+  | custreamz/custreamz/tests/
+  | dask_cudf/dask_cudf/tests/
+  # This close paren cannot be in column zero otherwise the config parser barfs
+ )