chore: Replace black by ruff format (#11996)

.github/workflows/lint-python.yml

@@ -30,9 +30,9 @@ jobs:
 
       - name: Lint Python
         run: |
-          ruff --exit-non-zero-on-fix .
-          black --check .
-          blackdoc --check .
+          ruff check --diff .
+          ruff format --diff .
+          blackdoc --diff .
 
   mypy:
     runs-on: ubuntu-latest

CONTRIBUTING.md

@@ -100,8 +100,8 @@ make pre-commit
 Note that we do not actually use the [pre-commit](https://pre-commit.com/) tool.
 We use the Makefile to conveniently run the following formatting and linting tools:
 
-- [black](https://black.readthedocs.io/) and [blackdoc](https://github.com/keewis/blackdoc)
 - [ruff](https://github.com/charliermarsh/ruff)
+- [blackdoc](https://github.com/keewis/blackdoc)
 - [mypy](http://mypy-lang.org/)
 - [rustfmt](https://github.com/rust-lang/rustfmt)
 - [clippy](https://doc.rust-lang.org/nightly/clippy/index.html)

docs/development/contributing.md

@@ -95,8 +95,8 @@ make pre-commit
 Note that we do not actually use the [pre-commit](https://pre-commit.com/) tool.
 We use the Makefile to conveniently run the following formatting and linting tools:
 
-- [black](https://black.readthedocs.io/) and [blackdoc](https://github.com/keewis/blackdoc)
 - [ruff](https://github.com/charliermarsh/ruff)
+- [blackdoc](https://github.com/keewis/blackdoc)
 - [mypy](http://mypy-lang.org/)
 - [rustfmt](https://github.com/rust-lang/rustfmt)
 - [clippy](https://doc.rust-lang.org/nightly/clippy/index.html)

py-polars/Makefile

@@ -56,8 +56,8 @@ build-release-native: .venv  ## Same as build-release, except with native CPU op
 
 .PHONY: fmt
 fmt: .venv  ## Run autoformatting and linting
-	$(VENV_BIN)/ruff .
-	$(VENV_BIN)/black .
+	$(VENV_BIN)/ruff check .
+	$(VENV_BIN)/ruff format .
 	$(VENV_BIN)/blackdoc .
 	$(VENV_BIN)/typos ..
 	cargo fmt --all

py-polars/polars/config.py

@@ -333,7 +333,9 @@ def save_to_file(cls, file: Path | str) -> None:
     @classmethod
     @deprecate_nonkeyword_arguments(version="0.19.3")
     def state(
-        cls, if_set: bool = False, env_only: bool = False  # noqa: FBT001
+        cls,
+        if_set: bool = False,  # noqa: FBT001
+        env_only: bool = False,  # noqa: FBT001
     ) -> dict[str, str | None]:
         """
         Show the current state of all Config variables as a dict.

py-polars/polars/dataframe/frame.py

@@ -23,6 +23,7 @@
     NoReturn,
     Sequence,
     TypeVar,
+    Union,
     cast,
     overload,
 )
@@ -174,10 +175,10 @@
     # MultiColSelector indexes into the horizontal axis
     # NOTE: wrapping these as strings is necessary for Python <3.10
 
-    MultiRowSelector: TypeAlias = "slice | range | list[int] | Series"
-    MultiColSelector: TypeAlias = (
-        "slice | range | list[int] | list[str] | list[bool] | Series"
-    )
+    MultiRowSelector: TypeAlias = Union[slice, range, "list[int]", "Series"]
+    MultiColSelector: TypeAlias = Union[
+        slice, range, "list[int]", "list[str]", "list[bool]", "Series"
+    ]
 
     T = TypeVar("T")
     P = ParamSpec("P")
@@ -1276,7 +1277,9 @@ def __array__(self, dtype: Any = None) -> np.ndarray[Any, Any]:
             return self.to_numpy().__array__()
 
     def __dataframe__(
-        self, nan_as_null: bool = False, allow_copy: bool = True  # noqa: FBT001
+        self,
+        nan_as_null: bool = False,  # noqa: FBT001
+        allow_copy: bool = True,  # noqa: FBT001
     ) -> PolarsDataFrame:
         """
         Convert to a dataframe object implementing the dataframe interchange protocol.
@@ -1911,13 +1914,15 @@ def to_dict(self, as_series: Literal[False]) -> dict[str, list[Any]]:
 
     @overload
     def to_dict(
-        self, as_series: bool  # noqa: FBT001
+        self,
+        as_series: bool,  # noqa: FBT001
     ) -> dict[str, Series] | dict[str, list[Any]]:
         ...
 
     # TODO: Make `as_series` keyword-only
     def to_dict(
-        self, as_series: bool = True  # noqa: FBT001
+        self,
+        as_series: bool = True,  # noqa: FBT001
     ) -> dict[str, Series] | dict[str, list[Any]]:
         """
         Convert DataFrame to a dictionary mapping column name to values.
@@ -2029,7 +2034,10 @@ def to_dicts(self) -> list[dict[str, Any]]:
 
     @deprecate_nonkeyword_arguments(version="0.19.3")
     def to_numpy(
-        self, structured: bool = False, *, order: IndexOrder = "fortran"  # noqa: FBT001
+        self,
+        structured: bool = False,  # noqa: FBT001
+        *,
+        order: IndexOrder = "fortran",
     ) -> np.ndarray[Any, Any]:
         """
         Convert DataFrame to a 2D NumPy array.
@@ -3094,7 +3102,11 @@ def write_excel(
                 options = {"hidden": True}
             if column in column_widths:  # type: ignore[operator]
                 ws.set_column_pixels(
-                    col_idx, col_idx, column_widths[column], None, options  # type: ignore[index]
+                    col_idx,
+                    col_idx,
+                    column_widths[column],  # type: ignore[index]
+                    None,
+                    options,
                 )
             elif options:
                 ws.set_column(col_idx, col_idx, None, None, options)

py-polars/polars/datatypes/convert.py

@@ -143,9 +143,7 @@ def _map_py_type_to_dtype(
             if len(nested) == 1:
                 nested = nested[0]
             return (
-                dtype
-                if nested is None
-                else dtype(_map_py_type_to_dtype(nested))  # type: ignore[operator]
+                dtype if nested is None else dtype(_map_py_type_to_dtype(nested))  # type: ignore[operator]
             )
 
     raise TypeError("invalid type")

py-polars/polars/interchange/dataframe.py

@@ -36,7 +36,9 @@ def __init__(self, df: DataFrame, *, allow_copy: bool = True):
         self._allow_copy = allow_copy
 
     def __dataframe__(
-        self, nan_as_null: bool = False, allow_copy: bool = True  # noqa: FBT001
+        self,
+        nan_as_null: bool = False,  # noqa: FBT001
+        allow_copy: bool = True,  # noqa: FBT001
     ) -> PolarsDataFrame:
         """
         Construct a new dataframe object, potentially changing the parameters.

py-polars/polars/interchange/protocol.py

@@ -193,7 +193,9 @@ def version(self) -> int:
         """Version of the protocol."""
 
     def __dataframe__(
-        self, nan_as_null: bool = False, allow_copy: bool = True  # noqa: FBT001
+        self,
+        nan_as_null: bool = False,  # noqa: FBT001
+        allow_copy: bool = True,  # noqa: FBT001
     ) -> DataFrame:
         """Convert to a dataframe object implementing the dataframe interchange protocol."""  # noqa: W505
 
@@ -236,7 +238,9 @@ class SupportsInterchange(Protocol):
     """Dataframe that supports conversion into an interchange dataframe object."""
 
     def __dataframe__(
-        self, nan_as_null: bool = False, allow_copy: bool = True  # noqa: FBT001
+        self,
+        nan_as_null: bool = False,  # noqa: FBT001
+        allow_copy: bool = True,  # noqa: FBT001
     ) -> SupportsInterchange:
         """Convert to a dataframe object implementing the dataframe interchange protocol."""  # noqa: W505
 

py-polars/polars/testing/parametric/primitives.py

@@ -718,9 +718,7 @@ def draw_frames(draw: DrawFn) -> DataFrame | LazyFrame:
                         schema={repr(schema).replace("', ","', pl.")},
                         orient={orient!r},
                     )
-                    """.replace(
-                        "datetime.", ""
-                    )
+                    """.replace("datetime.", "")
                 )
                 # note: this avoids printing the repro twice
                 if failed_frame_init not in _failed_frame_init_msgs_:

py-polars/polars/utils/_async.py

@@ -34,7 +34,9 @@ def __init__(self) -> None:
         self._watcher.start(self._watcher_callback)
 
     def get(
-        self, block: bool = True, timeout: float | int | None = None  # noqa: FBT001
+        self,
+        block: bool = True,  # noqa: FBT001
+        timeout: float | int | None = None,
     ) -> T:
         return self.result.get(block=block, timeout=timeout)
 

py-polars/polars/utils/_construction.py

@@ -352,8 +352,7 @@ def _construct_series_with_fallbacks(
             if "'float'" in str_exc and (
                 # we do not accept float values as int/temporal, as it causes silent
                 # information loss; the caller should explicitly cast in this case.
-                target_dtype
-                not in (INTEGER_DTYPES | TEMPORAL_DTYPES)
+                target_dtype not in (INTEGER_DTYPES | TEMPORAL_DTYPES)
             ):
                 constructor = py_type_to_constructor(float)
 
@@ -741,9 +740,7 @@ def _unpack_schema(
         else None
     )
     column_dtypes: dict[str, PolarsDataType] = {
-        lookup.get((name := col[0]), name)
-        if lookup
-        else col[0]: dtype  # type: ignore[misc]
+        lookup.get((name := col[0]), name) if lookup else col[0]: dtype  # type: ignore[misc]
         if is_polars_dtype(dtype, include_unknown=True)
         else py_type_to_dtype(dtype)
         for col in schema

py-polars/pyproject.toml

@@ -130,7 +130,6 @@ select = [
   "SIM", # flake8-simplify
   "TCH", # flake8-type-checking
   "TID", # flake8-tidy-imports
-  "Q", # flake8-quotes
   "UP", # pyupgrade
   "PT", # flake8-pytest-style
   "RUF", # Ruff-specific rules
@@ -172,14 +171,14 @@ ignore = [
   "TD003", # Missing issue link on the line following this TODO
   # tryceratops
   "TRY003", # Avoid specifying long messages outside the exception class
+  # Lints below are turned off because of conflicts with the ruff formatter
+  "D206",
+  "W191",
 ]
 
 [tool.ruff.pycodestyle]
 max-doc-length = 88
 
-[tool.ruff.isort]
-split-on-trailing-comma = false
-
 [tool.ruff.flake8-tidy-imports]
 ban-relative-imports = "all"
 

py-polars/requirements-lint.txt

@@ -1,5 +1,4 @@
-black==23.10.0
 blackdoc==0.3.8
 mypy==1.6.0
-ruff==0.1.0
+ruff==0.1.2
 typos==1.16.20

py-polars/tests/unit/dataframe/test_df.py

@@ -2641,9 +2641,7 @@ def test_fill_null_limits() -> None:
             pl.all().fill_null(strategy="forward", limit=2),
             pl.all().fill_null(strategy="backward", limit=2).name.suffix("_backward"),
         ]
-    ).to_dict(
-        False
-    ) == {
+    ).to_dict(False) == {
         "a": [1, 1, 1, None, 5, 6, 6, 6, None, 10],
         "b": ["a", "a", "a", None, "b", "c", "c", "c", None, "d"],
         "c": [True, True, True, None, False, True, True, True, None, False],
@@ -2774,15 +2772,17 @@ def test_selection_regex_and_multicol() -> None:
         ]
 
 
-def test_unique_on_sorted() -> None:
+@pytest.mark.parametrize("subset", ["a", cs.starts_with("x", "a")])
+def test_unique_on_sorted(subset: Any) -> None:
     df = pl.DataFrame(data={"a": [1, 1, 3], "b": [1, 2, 3]})
-    for subset in ("a", cs.starts_with("x", "a")):
-        assert df.with_columns([pl.col("a").set_sorted()]).unique(
-            subset=subset, keep="last"  # type: ignore[arg-type]
-        ).to_dict(False) == {
-            "a": [1, 3],
-            "b": [2, 3],
-        }
+
+    result = df.with_columns([pl.col("a").set_sorted()]).unique(
+        subset=subset,
+        keep="last",
+    )
+
+    expected = pl.DataFrame({"a": [1, 3], "b": [2, 3]})
+    assert_frame_equal(result, expected)
 
 
 def test_len_compute(df: pl.DataFrame) -> None:

py-polars/tests/unit/datatypes/test_categorical.py

@@ -299,15 +299,9 @@ def test_nested_categorical_aggregation_7848() -> None:
         }
     ).with_columns([pl.col("letter").cast(pl.Categorical)]).group_by(
         maintain_order=True, by=["group"]
-    ).all().with_columns(
-        [pl.col("letter").list.len().alias("c_group")]
-    ).group_by(
+    ).all().with_columns(pl.col("letter").list.len().alias("c_group")).group_by(
         by=["c_group"], maintain_order=True
-    ).agg(
-        pl.col("letter")
-    ).to_dict(
-        False
-    ) == {
+    ).agg(pl.col("letter")).to_dict(False) == {
         "c_group": [2, 3],
         "letter": [[["a", "b"], ["f", "g"]], [["c", "d", "e"]]],
     }

py-polars/tests/unit/datatypes/test_decimal.py

@@ -227,6 +227,8 @@ def test_decimal_aggregations() -> None:
         sum=pl.sum("a"),
         min=pl.min("a"),
         max=pl.max("a"),
-    ).to_dict(
-        False
-    ) == {"sum": [D("9110.33")], "min": [D("0.10")], "max": [D("9000.12")]}
+    ).to_dict(False) == {
+        "sum": [D("9110.33")],
+        "min": [D("0.10")],
+        "max": [D("9000.12")],
+    }

py-polars/tests/unit/datatypes/test_list.py

@@ -286,9 +286,7 @@ def test_list_count_matches_deprecated() -> None:
             {"listcol": [[], [1], [1, 2, 3, 2], [1, 2, 1], [4, 4]]}
         ).select(pl.col("listcol").list.count_match(2).alias("number_of_twos")).to_dict(
             False
-        ) == {
-            "number_of_twos": [0, 0, 2, 1, 0]
-        }
+        ) == {"number_of_twos": [0, 0, 2, 1, 0]}
 
 
 def test_list_count_matches() -> None:

py-polars/tests/unit/datatypes/test_struct.py

@@ -317,9 +317,7 @@ def test_struct_list_head_tail() -> None:
             pl.col("list_of_struct").list.head(1).alias("head"),
             pl.col("list_of_struct").list.tail(1).alias("tail"),
         ]
-    ).to_dict(
-        False
-    ) == {
+    ).to_dict(False) == {
         "list_of_struct": [
             [{"a": 1, "b": 4}, {"a": 3, "b": 6}],
             [{"a": 10, "b": 40}, {"a": 20, "b": 50}, {"a": 30, "b": 60}],
@@ -401,9 +399,7 @@ def test_struct_concat_list() -> None:
         }
     ).with_columns(
         [pl.col("list_struct1").list.concat("list_struct2").alias("result")]
-    )[
-        "result"
-    ].to_list() == [
+    )["result"].to_list() == [
         [{"a": 1, "b": 2}, {"a": 3, "b": 4}, {"a": 6, "b": 7}, {"a": 8, "b": 9}],
         [{"a": 1, "b": 2}, {"a": 6, "b": 7}],
     ]

py-polars/tests/unit/datatypes/test_temporal.py

@@ -1263,11 +1263,7 @@ def test_unique_counts_on_dates() -> None:
             pl.col("dt_ns").dt.cast_time_unit("ms").alias("dt_ms"),
             pl.col("dt_ns").cast(pl.Date).alias("date"),
         ]
-    ).select(
-        pl.all().unique_counts().sum()
-    ).to_dict(
-        False
-    ) == {
+    ).select(pl.all().unique_counts().sum()).to_dict(False) == {
         "dt_ns": [3],
         "dt_us": [3],
         "dt_ms": [3],
@@ -1304,9 +1300,7 @@ def test_rolling_by_ordering() -> None:
         [
             pl.col("val").sum().alias("sum val"),
         ]
-    ).to_dict(
-        False
-    ) == {
+    ).to_dict(False) == {
         "key": ["A", "A", "A", "A", "B", "B", "B"],
         "dt": [
             datetime(2022, 1, 1, 0, 1),
@@ -1402,9 +1396,7 @@ def test_sum_duration() -> None:
         ]
     ).select(
         [pl.col("duration").sum(), pl.col("duration").dt.seconds().alias("sec").sum()]
-    ).to_dict(
-        False
-    ) == {
+    ).to_dict(False) == {
         "duration": [timedelta(seconds=150)],
         "sec": [150],
     }