TYP: some return types from ruff (#56617)

* TYP: some return types from ruff

* fix CI

* and another one

* adjust pre-commit

.pre-commit-config.yaml

@@ -32,6 +32,8 @@ repos:
         # TODO: remove autofixe-only rules when they are checked by ruff
         name: ruff-selected-autofixes
         alias: ruff-selected-autofixes
+        files: ^pandas
+        exclude: ^pandas/tests
         args: [--select, "ANN001,ANN2", --fix-only, --exit-non-zero-on-fix]
 -   repo: https://github.com/jendrikseipp/vulture
     rev: 'v2.10'

doc/source/whatsnew/v2.2.0.rst

@@ -431,7 +431,7 @@ Optional libraries below the lowest tested version may still work, but are not c
 +-----------------+-----------------+---------+
 | Package         | Minimum Version | Changed |
 +=================+=================+=========+
-| mypy (dev)      | 1.7.1           |    X    |
+| mypy (dev)      | 1.8.0           |    X    |
 +-----------------+-----------------+---------+
 |                 |                 |    X    |
 +-----------------+-----------------+---------+

environment.yml

@@ -76,7 +76,7 @@ dependencies:
 
   # code checks
   - flake8=6.1.0  # run in subprocess over docstring examples
-  - mypy=1.7.1  # pre-commit uses locally installed mypy
+  - mypy=1.8.0  # pre-commit uses locally installed mypy
   - tokenize-rt  # scripts/check_for_inconsistent_pandas_namespace.py
   - pre-commit>=3.6.0
 

pandas/_testing/asserters.py

@@ -4,6 +4,7 @@
 from typing import (
     TYPE_CHECKING,
     Literal,
+    NoReturn,
     cast,
 )
 
@@ -143,7 +144,7 @@ def assert_almost_equal(
         )
 
 
-def _check_isinstance(left, right, cls):
+def _check_isinstance(left, right, cls) -> None:
     """
     Helper method for our assert_* methods that ensures that
     the two objects being compared have the right type before
@@ -576,7 +577,7 @@ def assert_timedelta_array_equal(
 
 def raise_assert_detail(
     obj, message, left, right, diff=None, first_diff=None, index_values=None
-):
+) -> NoReturn:
     __tracebackhide__ = True
 
     msg = f"""{obj} are different
@@ -664,7 +665,7 @@ def _get_base(obj):
         if left_base is right_base:
             raise AssertionError(f"{repr(left_base)} is {repr(right_base)}")
 
-    def _raise(left, right, err_msg):
+    def _raise(left, right, err_msg) -> NoReturn:
         if err_msg is None:
             if left.shape != right.shape:
                 raise_assert_detail(

pandas/_version.py

@@ -386,7 +386,7 @@ def git_pieces_from_vcs(tag_prefix, root, verbose, runner=run_command):
     return pieces
 
 
-def plus_or_dot(pieces):
+def plus_or_dot(pieces) -> str:
     """Return a + if we don't already have one, else return a ."""
     if "+" in pieces.get("closest-tag", ""):
         return "."

pandas/core/computation/expr.py

@@ -695,8 +695,8 @@ def visit_Call(self, node, side=None, **kwargs):
                 if not isinstance(key, ast.keyword):
                     # error: "expr" has no attribute "id"
                     raise ValueError(
-                        "keyword error in function call "  # type: ignore[attr-defined]
-                        f"'{node.func.id}'"
+                        "keyword error in function call "
+                        f"'{node.func.id}'"  # type: ignore[attr-defined]
                     )
 
                 if key.arg:

pandas/io/html.py

@@ -269,7 +269,7 @@ def _attr_getter(self, obj, attr):
         # Both lxml and BeautifulSoup have the same implementation:
         return obj.get(attr)
 
-    def _href_getter(self, obj):
+    def _href_getter(self, obj) -> str | None:
         """
         Return a href if the DOM node contains a child <a> or None.
 
@@ -392,7 +392,7 @@ def _parse_tables(self, document, match, attrs):
         """
         raise AbstractMethodError(self)
 
-    def _equals_tag(self, obj, tag):
+    def _equals_tag(self, obj, tag) -> bool:
         """
         Return whether an individual DOM node matches a tag
 
@@ -629,7 +629,7 @@ def _href_getter(self, obj) -> str | None:
     def _text_getter(self, obj):
         return obj.text
 
-    def _equals_tag(self, obj, tag):
+    def _equals_tag(self, obj, tag) -> bool:
         return obj.name == tag
 
     def _parse_td(self, row):
@@ -758,7 +758,7 @@ def _parse_tables(self, document, match, kwargs):
             raise ValueError(f"No tables found matching regex {repr(pattern)}")
         return tables
 
-    def _equals_tag(self, obj, tag):
+    def _equals_tag(self, obj, tag) -> bool:
         return obj.tag == tag
 
     def _build_doc(self):

pandas/io/json/_json.py

@@ -255,7 +255,7 @@ def __init__(
         self.is_copy = None
         self._format_axes()
 
-    def _format_axes(self):
+    def _format_axes(self) -> None:
         raise AbstractMethodError(self)
 
     def write(self) -> str:
@@ -287,7 +287,7 @@ def obj_to_write(self) -> NDFrame | Mapping[IndexLabel, Any]:
         else:
             return self.obj
 
-    def _format_axes(self):
+    def _format_axes(self) -> None:
         if not self.obj.index.is_unique and self.orient == "index":
             raise ValueError(f"Series index must be unique for orient='{self.orient}'")
 
@@ -304,7 +304,7 @@ def obj_to_write(self) -> NDFrame | Mapping[IndexLabel, Any]:
             obj_to_write = self.obj
         return obj_to_write
 
-    def _format_axes(self):
+    def _format_axes(self) -> None:
         """
         Try to format axes if they are datelike.
         """
@@ -1193,7 +1193,7 @@ def parse(self):
         self._try_convert_types()
         return self.obj
 
-    def _parse(self):
+    def _parse(self) -> None:
         raise AbstractMethodError(self)
 
     @final
@@ -1217,7 +1217,7 @@ def _convert_axes(self) -> None:
                 new_axis = Index(new_ser, dtype=new_ser.dtype, copy=False)
                 setattr(self.obj, axis_name, new_axis)
 
-    def _try_convert_types(self):
+    def _try_convert_types(self) -> None:
         raise AbstractMethodError(self)
 
     @final

pandas/io/parsers/arrow_parser_wrapper.py

@@ -41,7 +41,7 @@ def __init__(self, src: ReadBuffer[bytes], **kwds) -> None:
 
         self._parse_kwds()
 
-    def _parse_kwds(self):
+    def _parse_kwds(self) -> None:
         """
         Validates keywords before passing to pyarrow.
         """
@@ -104,7 +104,7 @@ def _get_pyarrow_options(self) -> None:
                 ] = None  # PyArrow raises an exception by default
             elif on_bad_lines == ParserBase.BadLineHandleMethod.WARN:
 
-                def handle_warning(invalid_row):
+                def handle_warning(invalid_row) -> str:
                     warnings.warn(
                         f"Expected {invalid_row.expected_columns} columns, but found "
                         f"{invalid_row.actual_columns}: {invalid_row.text}",
@@ -219,7 +219,7 @@ def _finalize_pandas_output(self, frame: DataFrame) -> DataFrame:
                 raise ValueError(e)
         return frame
 
-    def _validate_usecols(self, usecols):
+    def _validate_usecols(self, usecols) -> None:
         if lib.is_list_like(usecols) and not all(isinstance(x, str) for x in usecols):
             raise ValueError(
                 "The pyarrow engine does not allow 'usecols' to be integer "

pandas/io/pytables.py

@@ -1707,7 +1707,7 @@ def info(self) -> str:
     # ------------------------------------------------------------------------
     # private methods
 
-    def _check_if_open(self):
+    def _check_if_open(self) -> None:
         if not self.is_open:
             raise ClosedFileError(f"{self._path} file is not open!")
 

pandas/io/sas/sas_xport.py

@@ -288,7 +288,7 @@ def close(self) -> None:
     def _get_row(self):
         return self.filepath_or_buffer.read(80).decode()
 
-    def _read_header(self):
+    def _read_header(self) -> None:
         self.filepath_or_buffer.seek(0)
 
         # read file header

pandas/io/sql.py

@@ -1514,7 +1514,7 @@ def _create_sql_schema(
         keys: list[str] | None = None,
         dtype: DtypeArg | None = None,
         schema: str | None = None,
-    ):
+    ) -> str:
         pass
 
 
@@ -2073,7 +2073,7 @@ def _create_sql_schema(
         keys: list[str] | None = None,
         dtype: DtypeArg | None = None,
         schema: str | None = None,
-    ):
+    ) -> str:
         table = SQLTable(
             table_name,
             self,
@@ -2433,7 +2433,7 @@ def _create_sql_schema(
         keys: list[str] | None = None,
         dtype: DtypeArg | None = None,
         schema: str | None = None,
-    ):
+    ) -> str:
         raise NotImplementedError("not implemented for adbc")
 
 
@@ -2879,7 +2879,7 @@ def _create_sql_schema(
         keys=None,
         dtype: DtypeArg | None = None,
         schema: str | None = None,
-    ):
+    ) -> str:
         table = SQLiteTable(
             table_name,
             self,

pandas/io/stata.py

@@ -687,7 +687,7 @@ def __init__(
 
         self._prepare_value_labels()
 
-    def _prepare_value_labels(self):
+    def _prepare_value_labels(self) -> None:
         """Encode value labels."""
 
         self.text_len = 0

pandas/plotting/_matplotlib/core.py

@@ -662,7 +662,7 @@ def _ensure_frame(self, data) -> DataFrame:
         return data
 
     @final
-    def _compute_plot_data(self):
+    def _compute_plot_data(self) -> None:
         data = self.data
 
         # GH15079 reconstruct data if by is defined
@@ -699,7 +699,7 @@ def _compute_plot_data(self):
 
         self.data = numeric_data.apply(type(self)._convert_to_ndarray)
 
-    def _make_plot(self, fig: Figure):
+    def _make_plot(self, fig: Figure) -> None:
         raise AbstractMethodError(self)
 
     @final
@@ -745,7 +745,7 @@ def _post_plot_logic(self, ax: Axes, data) -> None:
         """Post process for each axes. Overridden in child classes"""
 
     @final
-    def _adorn_subplots(self, fig: Figure):
+    def _adorn_subplots(self, fig: Figure) -> None:
         """Common post process unrelated to data"""
         if len(self.axes) > 0:
             all_axes = self._get_subplots(fig)
@@ -1323,7 +1323,7 @@ def __init__(
             c = self.data.columns[c]
         self.c = c
 
-    def _make_plot(self, fig: Figure):
+    def _make_plot(self, fig: Figure) -> None:
         x, y, c, data = self.x, self.y, self.c, self.data
         ax = self.axes[0]
 

pandas/util/_validators.py

@@ -26,7 +26,7 @@
 BoolishNoneT = TypeVar("BoolishNoneT", bool, int, None)
 
 
-def _check_arg_length(fname, args, max_fname_arg_count, compat_args):
+def _check_arg_length(fname, args, max_fname_arg_count, compat_args) -> None:
     """
     Checks whether 'args' has length of at most 'compat_args'. Raises
     a TypeError if that is not the case, similar to in Python when a
@@ -46,7 +46,7 @@ def _check_arg_length(fname, args, max_fname_arg_count, compat_args):
         )
 
 
-def _check_for_default_values(fname, arg_val_dict, compat_args):
+def _check_for_default_values(fname, arg_val_dict, compat_args) -> None:
     """
     Check that the keys in `arg_val_dict` are mapped to their
     default values as specified in `compat_args`.
@@ -125,7 +125,7 @@ def validate_args(fname, args, max_fname_arg_count, compat_args) -> None:
     _check_for_default_values(fname, kwargs, compat_args)
 
 
-def _check_for_invalid_keys(fname, kwargs, compat_args):
+def _check_for_invalid_keys(fname, kwargs, compat_args) -> None:
     """
     Checks whether 'kwargs' contains any keys that are not
     in 'compat_args' and raises a TypeError if there is one.

requirements-dev.txt

@@ -53,7 +53,7 @@ moto
 flask
 asv>=0.6.1
 flake8==6.1.0
-mypy==1.7.1
+mypy==1.8.0
 tokenize-rt
 pre-commit>=3.6.0
 gitpython