Add pyupgrade onto pre-commit (#6152)

* Add pyupgrade onto pre-commit

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Update doc/whats-new.rst

Co-authored-by: Illviljan <[email protected]>

* .

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Illviljan <[email protected]>

Author: 3 people

.pre-commit-config.yaml

@@ -6,11 +6,24 @@ repos:
       - id: trailing-whitespace
       - id: end-of-file-fixer
       - id: check-yaml
-  # isort should run before black as black sometimes tweaks the isort output
+      - id: debug-statements
+      - id: mixed-line-ending
+    # This wants to go before isort & flake8
+  - repo: https://github.com/myint/autoflake
+    rev: "v1.4"
+    hooks:
+      - id: autoflake # isort should run before black as black sometimes tweaks the isort output
+        args: ["--in-place", "--ignore-init-module-imports"]
   - repo: https://github.com/PyCQA/isort
     rev: 5.10.1
     hooks:
       - id: isort
+  - repo: https://github.com/asottile/pyupgrade
+    rev: v2.31.0
+    hooks:
+      - id: pyupgrade
+        args:
+          - "--py37-plus"
   # https://github.com/python/black#version-control-integration
   - repo: https://github.com/psf/black
     rev: 21.12b0
@@ -47,12 +60,3 @@ repos:
             typing-extensions==3.10.0.0,
             numpy,
           ]
-  # run this occasionally, ref discussion https://github.com/pydata/xarray/pull/3194
-  # - repo: https://github.com/asottile/pyupgrade
-  #   rev: v1.22.1
-  #   hooks:
-  #     - id: pyupgrade
-  #       args:
-  #         - "--py3-only"
-  #         # remove on f-strings in Py3.7
-  #         - "--keep-percent-format"

doc/whats-new.rst

@@ -78,6 +78,8 @@ Internal Changes
 - Removed internal checks for ``pd.Panel`` (:issue:`6145`).
   By `Matthew Roeschke <https://github.com/mroeschke>`_.
 
+- Add ``pyupgrade`` pre-commit hook (:pull:`6152`).
+  By `Maximilian Roos <https://github.com/max-sixty>`_.
 
 .. _whats-new.0.20.2:
 

xarray/backends/plugins.py

@@ -168,10 +168,8 @@ def get_backend(engine):
         backend = engine()
     else:
         raise TypeError(
-            (
-                "engine must be a string or a subclass of "
-                f"xarray.backends.BackendEntrypoint: {engine}"
-            )
+            "engine must be a string or a subclass of "
+            f"xarray.backends.BackendEntrypoint: {engine}"
         )
 
     return backend

xarray/coding/cftime_offsets.py

@@ -160,7 +160,7 @@ def rollback(self, date):
             return date - type(self)()
 
     def __str__(self):
-        return "<{}: n={}>".format(type(self).__name__, self.n)
+        return f"<{type(self).__name__}: n={self.n}>"
 
     def __repr__(self):
         return str(self)
@@ -399,10 +399,10 @@ def __mul__(self, other):
         return type(self)(n=other * self.n, month=self.month)
 
     def rule_code(self):
-        return "{}-{}".format(self._freq, _MONTH_ABBREVIATIONS[self.month])
+        return f"{self._freq}-{_MONTH_ABBREVIATIONS[self.month]}"
 
     def __str__(self):
-        return "<{}: n={}, month={}>".format(type(self).__name__, self.n, self.month)
+        return f"<{type(self).__name__}: n={self.n}, month={self.month}>"
 
 
 class QuarterBegin(QuarterOffset):
@@ -485,10 +485,10 @@ def __mul__(self, other):
         return type(self)(n=other * self.n, month=self.month)
 
     def rule_code(self):
-        return "{}-{}".format(self._freq, _MONTH_ABBREVIATIONS[self.month])
+        return f"{self._freq}-{_MONTH_ABBREVIATIONS[self.month]}"
 
     def __str__(self):
-        return "<{}: n={}, month={}>".format(type(self).__name__, self.n, self.month)
+        return f"<{type(self).__name__}: n={self.n}, month={self.month}>"
 
 
 class YearBegin(YearOffset):
@@ -741,7 +741,7 @@ def _generate_linear_range(start, end, periods):
 
     total_seconds = (end - start).total_seconds()
     values = np.linspace(0.0, total_seconds, periods, endpoint=True)
-    units = "seconds since {}".format(format_cftime_datetime(start))
+    units = f"seconds since {format_cftime_datetime(start)}"
     calendar = start.calendar
     return cftime.num2date(
         values, units=units, calendar=calendar, only_use_cftime_datetimes=True

xarray/coding/strings.py

@@ -18,7 +18,7 @@
 
 def create_vlen_dtype(element_type):
     if element_type not in (str, bytes):
-        raise TypeError("unsupported type for vlen_dtype: {!r}".format(element_type))
+        raise TypeError(f"unsupported type for vlen_dtype: {element_type!r}")
     # based on h5py.special_dtype
     return np.dtype("O", metadata={"element_type": element_type})
 
@@ -227,7 +227,7 @@ def shape(self):
         return self.array.shape[:-1]
 
     def __repr__(self):
-        return "{}({!r})".format(type(self).__name__, self.array)
+        return f"{type(self).__name__}({self.array!r})"
 
     def __getitem__(self, key):
         # require slicing the last dimension completely

xarray/coding/times.py

@@ -131,8 +131,8 @@ def _ensure_padded_year(ref_date):
     matches_start_digits = re.match(r"(\d+)(.*)", ref_date)
     if not matches_start_digits:
         raise ValueError(f"invalid reference date for time units: {ref_date}")
-    ref_year, everything_else = [s for s in matches_start_digits.groups()]
-    ref_date_padded = "{:04d}{}".format(int(ref_year), everything_else)
+    ref_year, everything_else = (s for s in matches_start_digits.groups())
+    ref_date_padded = f"{int(ref_year):04d}{everything_else}"
 
     warning_msg = (
         f"Ambiguous reference date string: {ref_date}. The first value is "
@@ -155,7 +155,7 @@ def _unpack_netcdf_time_units(units):
     if not matches:
         raise ValueError(f"invalid time units: {units}")
 
-    delta_units, ref_date = [s.strip() for s in matches.groups()]
+    delta_units, ref_date = (s.strip() for s in matches.groups())
     ref_date = _ensure_padded_year(ref_date)
 
     return delta_units, ref_date
@@ -545,7 +545,7 @@ def _should_cftime_be_used(source, target_calendar, use_cftime):
 def _cleanup_netcdf_time_units(units):
     delta, ref_date = _unpack_netcdf_time_units(units)
     try:
-        units = "{} since {}".format(delta, format_timestamp(ref_date))
+        units = f"{delta} since {format_timestamp(ref_date)}"
     except (OutOfBoundsDatetime, ValueError):
         # don't worry about reifying the units if they're out of bounds or
         # formatted badly

xarray/core/common.py

@@ -8,16 +8,11 @@
     TYPE_CHECKING,
     Any,
     Callable,
-    Dict,
     Hashable,
     Iterable,
     Iterator,
-    List,
     Mapping,
-    Optional,
-    Tuple,
     TypeVar,
-    Union,
     overload,
 )
 
@@ -164,9 +159,7 @@ def __iter__(self: Any) -> Iterator[Any]:
             raise TypeError("iteration over a 0-d array")
         return self._iter()
 
-    def get_axis_num(
-        self, dim: Union[Hashable, Iterable[Hashable]]
-    ) -> Union[int, Tuple[int, ...]]:
+    def get_axis_num(self, dim: Hashable | Iterable[Hashable]) -> int | tuple[int, ...]:
         """Return axis number(s) corresponding to dimension(s) in this array.
 
         Parameters
@@ -244,7 +237,7 @@ def __getattr__(self, name: str) -> Any:
                 with suppress(KeyError):
                     return source[name]
         raise AttributeError(
-            "{!r} object has no attribute {!r}".format(type(self).__name__, name)
+            f"{type(self).__name__!r} object has no attribute {name!r}"
         )
 
     # This complicated two-method design boosts overall performance of simple operations
@@ -284,37 +277,37 @@ def __setattr__(self, name: str, value: Any) -> None:
                 "assignment (e.g., `ds['name'] = ...`) instead of assigning variables."
             ) from e
 
-    def __dir__(self) -> List[str]:
+    def __dir__(self) -> list[str]:
         """Provide method name lookup and completion. Only provide 'public'
         methods.
         """
-        extra_attrs = set(
+        extra_attrs = {
             item
             for source in self._attr_sources
             for item in source
             if isinstance(item, str)
-        )
+        }
         return sorted(set(dir(type(self))) | extra_attrs)
 
-    def _ipython_key_completions_(self) -> List[str]:
+    def _ipython_key_completions_(self) -> list[str]:
         """Provide method for the key-autocompletions in IPython.
         See http://ipython.readthedocs.io/en/stable/config/integrating.html#tab-completion
         For the details.
         """
-        items = set(
+        items = {
             item
             for source in self._item_sources
             for item in source
             if isinstance(item, str)
-        )
+        }
         return list(items)
 
 
 def get_squeeze_dims(
     xarray_obj,
-    dim: Union[Hashable, Iterable[Hashable], None] = None,
-    axis: Union[int, Iterable[int], None] = None,
-) -> List[Hashable]:
+    dim: Hashable | Iterable[Hashable] | None = None,
+    axis: int | Iterable[int] | None = None,
+) -> list[Hashable]:
     """Get a list of dimensions to squeeze out."""
     if dim is not None and axis is not None:
         raise ValueError("cannot use both parameters `axis` and `dim`")
@@ -346,15 +339,15 @@ def get_squeeze_dims(
 class DataWithCoords(AttrAccessMixin):
     """Shared base class for Dataset and DataArray."""
 
-    _close: Optional[Callable[[], None]]
+    _close: Callable[[], None] | None
 
     __slots__ = ("_close",)
 
     def squeeze(
         self,
-        dim: Union[Hashable, Iterable[Hashable], None] = None,
+        dim: Hashable | Iterable[Hashable] | None = None,
         drop: bool = False,
-        axis: Union[int, Iterable[int], None] = None,
+        axis: int | Iterable[int] | None = None,
     ):
         """Return a new object with squeezed data.
 
@@ -416,8 +409,8 @@ def get_index(self, key: Hashable) -> pd.Index:
             return pd.Index(range(self.sizes[key]), name=key)
 
     def _calc_assign_results(
-        self: C, kwargs: Mapping[Any, Union[T, Callable[[C], T]]]
-    ) -> Dict[Hashable, T]:
+        self: C, kwargs: Mapping[Any, T | Callable[[C], T]]
+    ) -> dict[Hashable, T]:
         return {k: v(self) if callable(v) else v for k, v in kwargs.items()}
 
     def assign_coords(self, coords=None, **coords_kwargs):
@@ -535,7 +528,7 @@ def assign_attrs(self, *args, **kwargs):
 
     def pipe(
         self,
-        func: Union[Callable[..., T], Tuple[Callable[..., T], str]],
+        func: Callable[..., T] | tuple[Callable[..., T], str],
         *args,
         **kwargs,
     ) -> T:
@@ -802,7 +795,7 @@ def groupby_bins(
             },
         )
 
-    def weighted(self: T_DataWithCoords, weights: "DataArray") -> Weighted[T_Xarray]:
+    def weighted(self: T_DataWithCoords, weights: DataArray) -> Weighted[T_Xarray]:
         """
         Weighted operations.
 
@@ -825,7 +818,7 @@ def rolling(
         self,
         dim: Mapping[Any, int] = None,
         min_periods: int = None,
-        center: Union[bool, Mapping[Any, bool]] = False,
+        center: bool | Mapping[Any, bool] = False,
         **window_kwargs: int,
     ):
         """
@@ -940,7 +933,7 @@ def coarsen(
         self,
         dim: Mapping[Any, int] = None,
         boundary: str = "exact",
-        side: Union[str, Mapping[Any, str]] = "left",
+        side: str | Mapping[Any, str] = "left",
         coord_func: str = "mean",
         **window_kwargs: int,
     ):
@@ -1290,7 +1283,7 @@ def where(self, cond, other=dtypes.NA, drop: bool = False):
 
         return ops.where_method(self, cond, other)
 
-    def set_close(self, close: Optional[Callable[[], None]]) -> None:
+    def set_close(self, close: Callable[[], None] | None) -> None:
         """Register the function that releases any resources linked to this object.
 
         This method controls how xarray cleans up resources associated
@@ -1523,20 +1516,20 @@ def __getitem__(self, value):
 
 @overload
 def full_like(
-    other: "Dataset",
+    other: Dataset,
     fill_value,
-    dtype: Union[DTypeLike, Mapping[Any, DTypeLike]] = None,
-) -> "Dataset":
+    dtype: DTypeLike | Mapping[Any, DTypeLike] = None,
+) -> Dataset:
     ...
 
 
 @overload
-def full_like(other: "DataArray", fill_value, dtype: DTypeLike = None) -> "DataArray":
+def full_like(other: DataArray, fill_value, dtype: DTypeLike = None) -> DataArray:
     ...
 
 
 @overload
-def full_like(other: "Variable", fill_value, dtype: DTypeLike = None) -> "Variable":
+def full_like(other: Variable, fill_value, dtype: DTypeLike = None) -> Variable:
     ...
 
 
@@ -1815,9 +1808,9 @@ def ones_like(other, dtype: DTypeLike = None):
 
 def get_chunksizes(
     variables: Iterable[Variable],
-) -> Mapping[Any, Tuple[int, ...]]:
+) -> Mapping[Any, tuple[int, ...]]:
 
-    chunks: Dict[Any, Tuple[int, ...]] = {}
+    chunks: dict[Any, tuple[int, ...]] = {}
     for v in variables:
         if hasattr(v.data, "chunks"):
             for dim, c in v.chunksizes.items():