flake8, bugbear, pyupgrade → ruff (#12002)

DimitriPapadopoulos · jacobtomlinson · web-flow · commit 17a245a70596 · 2025-10-31T14:44:16.000Z
Co-authored-by: Jacob Tomlinson &lt;jacobtomlinson@users.noreply.github.com&gt;
diff --git a/.flake8 b/.flake8
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
@@ -9,33 +9,22 @@ repos:
     hooks:
       - id: absolufy-imports
         name: absolufy-imports
+  - repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: v0.14.2
+    hooks:
+      - id: ruff-check
   - repo: https://github.com/pycqa/isort
     rev: 6.0.1
     hooks:
       - id: isort
         language_version: python3
-  - repo: https://github.com/asottile/pyupgrade
-    rev: v3.19.1
-    hooks:
-      - id: pyupgrade
-        args:
-          - --py310-plus
   - repo: https://github.com/psf/black
     rev: 25.1.0
     hooks:
       - id: black
         language_version: python3
         args:
           - --target-version=py310
-  - repo: https://github.com/pycqa/flake8
-    rev: 7.2.0
-    hooks:
-      - id: flake8
-        language_version: python3
-        additional_dependencies:
-          # NOTE: autoupdate does not pick up flake8-bugbear since it is a transitive
-          #  dependency. Make sure to update flake8-bugbear manually on a regular basis.
-          - flake8-bugbear==24.8.19
   - repo: https://github.com/codespell-project/codespell
     rev: v2.4.1
     hooks:
diff --git a/dask/array/linalg.py b/dask/array/linalg.py
@@ -1256,7 +1256,7 @@ def solve(a, b, sym_pos=None, assume_a="gen"):
         b = p.T.dot(b)
     else:
         raise ValueError(
-            f"{assume_a = } is not a recognized matrix structure, "  # noqa: E251
+            f"{assume_a = } is not a recognized matrix structure, "
             "valid structures in Dask are 'pos' and 'gen'."
         )
 
diff --git a/dask/array/numpy_compat.py b/dask/array/numpy_compat.py
@@ -18,7 +18,7 @@
 
 
 if NUMPY_GE_200:
-    from numpy.exceptions import AxisError, ComplexWarning  # noqa: F401
+    from numpy.exceptions import AxisError, ComplexWarning
     from numpy.lib.array_utils import normalize_axis_index, normalize_axis_tuple
 else:
     from numpy import (  # type: ignore[no-redef, attr-defined] # noqa: F401
diff --git a/dask/array/routines.py b/dask/array/routines.py
@@ -29,7 +29,7 @@
     tensordot_lookup,
 )
 from dask.array.creation import arange, diag, empty, indices, tri
-from dask.array.einsumfuncs import einsum  # noqa
+from dask.array.einsumfuncs import einsum  # noqa: F401
 from dask.array.numpy_compat import NUMPY_GE_200
 from dask.array.reductions import reduction
 from dask.array.ufunc import multiply, sqrt, true_divide
@@ -2086,7 +2086,7 @@ def _asarray_isnull(values):
 def isnull(values):
     """pandas.isnull for dask arrays"""
     # eagerly raise ImportError, if pandas isn't available
-    import pandas as pd  # noqa
+    import pandas as pd  # noqa: F401
 
     return elemwise(_asarray_isnull, values, dtype="bool")
 
diff --git a/dask/array/stats.py b/dask/array/stats.py
@@ -242,7 +242,7 @@ def skew(a, axis=0, bias=True, nan_policy="propagate"):
             "`nan_policy` other than 'propagate' have not been implemented."
         )
 
-    n = a.shape[axis]  # noqa; for bias
+    n = a.shape[axis]  # noqa: F841  # for bias
     m2 = moment(a, 2, axis)
     m3 = moment(a, 3, axis)
     zero = m2 == 0
@@ -298,7 +298,7 @@ def kurtosis(a, axis=0, fisher=True, bias=True, nan_policy="propagate"):
         raise NotImplementedError(
             "`nan_policy` other than 'propagate' have not been implemented."
         )
-    n = a.shape[axis]  # noqa; for bias
+    n = a.shape[axis]  # noqa: F841  # for bias
     m2 = moment(a, 2, axis)
     m4 = moment(a, 4, axis)
     zero = m2 == 0
diff --git a/dask/dataframe/__init__.py b/dask/dataframe/__init__.py
@@ -14,7 +14,7 @@ def _dask_expr_enabled() -> bool:
 _dask_expr_enabled()
 
 
-import dask.array._array_expr._backends  # Import this to register array dispatch # noqa: F401
+import dask.array._array_expr._backends  # Import this to register array dispatch
 
 # Ensure that dtypes are registered
 import dask.dataframe._dtypes
diff --git a/dask/dataframe/dask_expr/_repartition.py b/dask/dataframe/dask_expr/_repartition.py
@@ -459,7 +459,7 @@ def _divisions(self):
 
     def _lower(self):
         # populate cache
-        self._mem_usage  # noqa
+        self._mem_usage
         return super()._lower()
 
     def _layer(self) -> dict:
diff --git a/dask/dataframe/dask_expr/io/parquet.py b/dask/dataframe/dask_expr/io/parquet.py
@@ -1843,9 +1843,7 @@ def _divisions_from_statistics(aggregated_stats, index_name):
             col_ix = ix
             break
     else:
-        raise ValueError(
-            f"Index column {index_name} not found in statistics"  # noqa: E713
-        )
+        raise ValueError(f"Index column {index_name} not found in statistics")
     last_max = None
     minmax = []
     for file_stats in aggregated_stats:
diff --git a/dask/dataframe/dask_expr/io/tests/test_distributed.py b/dask/dataframe/dask_expr/io/tests/test_distributed.py
@@ -6,7 +6,7 @@
 
 distributed = pytest.importorskip("distributed")
 
-from distributed.utils_test import *  # noqa F401, F403
+from distributed.utils_test import *  # noqa: F401, F403
 from distributed.utils_test import gen_cluster
 
 import dask.dataframe as dd
diff --git a/dask/dataframe/dask_expr/tests/test_diagnostics.py b/dask/dataframe/dask_expr/tests/test_diagnostics.py
@@ -6,7 +6,7 @@
 
 pytest.importorskip("distributed")
 
-from distributed.utils_test import *  # noqa
+from distributed.utils_test import *  # noqa: F401, F403
 
 from dask.dataframe.dask_expr import from_pandas
 from dask.dataframe.dask_expr.tests._util import _backend_library
diff --git a/dask/dataframe/dask_expr/tests/test_distributed.py b/dask/dataframe/dask_expr/tests/test_distributed.py
@@ -13,7 +13,7 @@
 
 from distributed import Client, LocalCluster
 from distributed.shuffle._core import id_from_key
-from distributed.utils_test import gen_cluster  # noqa F401
+from distributed.utils_test import gen_cluster
 
 import dask.dataframe as dd
 
diff --git a/dask/dataframe/dask_expr/tests/test_format.py b/dask/dataframe/dask_expr/tests/test_format.py
@@ -1,4 +1,3 @@
-# flake8: noqa: W291
 from __future__ import annotations
 
 from textwrap import dedent
@@ -78,7 +77,7 @@ def test_to_string():
       <td>...</td>
     </tr>
   </tbody>
-</table>"""  # noqa E222, E702
+</table>"""
     footer = f"Dask Name: frompandas, {maybe_pluralize(1, 'expression')}"
     exp = f"""<div><strong>Dask DataFrame Structure:</strong></div>
 {exp_table}
diff --git a/dask/dataframe/tests/test_groupby.py b/dask/dataframe/tests/test_groupby.py
@@ -2521,7 +2521,7 @@ def test_groupby_grouper_dispatch(key):
     cudf = pytest.importorskip("cudf")
 
     # not directly used but must be imported
-    pytest.importorskip("dask_cudf")  # noqa: F841
+    pytest.importorskip("dask_cudf")
 
     pdf = pd.DataFrame(
         {
@@ -2555,7 +2555,7 @@ def test_groupby_apply_cudf(group_keys):
     # implied shuffle works for the `cudf` backend
 
     # Make sure test is skipped without dask_cudf
-    pytest.importorskip("dask_cudf")  # noqa: F841
+    pytest.importorskip("dask_cudf")
     cudf = pytest.importorskip("cudf")
 
     df = pd.DataFrame({"a": [1, 2, 3, 1, 2, 3], "b": [4, 5, 6, 7, 8, 9]})
diff --git a/dask/dataframe/utils.py b/dask/dataframe/utils.py
@@ -18,7 +18,7 @@
 from dask.base import is_dask_collection
 from dask.core import get_deps
 from dask.dataframe._compat import PANDAS_GE_300, tm  # noqa: F401
-from dask.dataframe.dispatch import (  # noqa : F401
+from dask.dataframe.dispatch import (  # noqa: F401
     is_categorical_dtype_dispatch,
     make_meta,
     make_meta_obj,
diff --git a/dask/distributed.py b/dask/distributed.py
@@ -1,5 +1,3 @@
-# flake8: noqa
-
 from __future__ import annotations
 
 _import_error_message = (
@@ -10,7 +8,7 @@
 )
 
 try:
-    from distributed import *
+    from distributed import *  # noqa: F403
 except ImportError as e:
     if e.msg == "No module named 'distributed'":
         raise ImportError(_import_error_message) from e
diff --git a/dask/local.py b/dask/local.py
@@ -259,7 +259,7 @@ def execute_task(key, task_info, dumps, loads, get_id, pack_exception):
         id = get_id()
         result = dumps((result, id))
         failed = False
-    except BaseException as e:  # noqa: B036
+    except BaseException as e:
         result = pack_exception(e, dumps)
         failed = True
     return key, result, failed
@@ -578,7 +578,7 @@ def submit(self, fn, *args, **kwargs):
         fut = Future()
         try:
             fut.set_result(fn(*args, **kwargs))
-        except BaseException as e:  # noqa: B036
+        except BaseException as e:
             fut.set_exception(e)
         return fut
 
diff --git a/dask/tests/test_distributed.py b/dask/tests/test_distributed.py
@@ -14,9 +14,9 @@
 from packaging.version import Version
 
 from distributed import Client, SchedulerPlugin, futures_of, wait
-from distributed.utils_test import cleanup  # noqa F401
-from distributed.utils_test import client as c  # noqa F401
-from distributed.utils_test import (  # noqa F401
+from distributed.utils_test import cleanup  # noqa: F401
+from distributed.utils_test import client as c  # noqa: F401
+from distributed.utils_test import (  # noqa: F401
     cluster,
     cluster_fixture,
     gen_cluster,
@@ -259,7 +259,7 @@ async def test_local_get_with_distributed_active(c, s, a, b):
     await asyncio.sleep(0.01)
     assert not s.tasks  # scheduler hasn't done anything
 
-    x = delayed(inc)(2).persist(scheduler="sync")  # noqa F841
+    x = delayed(inc)(2).persist(scheduler="sync")  # noqa: F841
     await asyncio.sleep(0.01)
     assert not s.tasks  # scheduler hasn't done anything
 
diff --git a/dask/tests/test_layers.py b/dask/tests/test_layers.py
@@ -10,7 +10,7 @@
 from operator import getitem
 
 from distributed import Client, SchedulerPlugin
-from distributed.utils_test import cluster, loop  # noqa F401
+from distributed.utils_test import cluster, loop  # noqa: F401
 
 from dask.layers import ArrayChunkShapeDep, ArraySliceDep, fractional_slice
 
diff --git a/dask/tests/test_ml.py b/dask/tests/test_ml.py
@@ -6,7 +6,7 @@ def test_basic():
         import dask_ml  # noqa: F401
     except ImportError:
         try:
-            from dask.ml.model_selection import GridSearchCV  # noqa: F401
+            from dask.ml.model_selection import GridSearchCV
         except ImportError as e:
             assert "conda install dask-ml" in str(e)
         else:
diff --git a/pyproject.toml b/pyproject.toml
@@ -93,6 +93,42 @@ find = {namespaces = false}
 [tool.setuptools.package-data]
 dask = ["py.typed"]
 
+[tool.ruff]
+line-length = 120
+
+[tool.ruff.lint]
+# By default, Ruff enables Flake8's F rules, along with a subset of the
+# E rules, omitting any stylistic rules that overlap with the use of a
+# formatter, like ruff format or Black. 
+# We specify additional rules with extend-select.
+extend-select = [
+    "B",
+    "UP",
+]
+ignore = [
+    "B011",   # Do not `assert False`, raise `AssertionError()`
+    "B018",   # Found useless expression
+    "B028",   # FIXME: No explicit `stacklevel` keyword argument found
+    "B904",   # Within an `except` clause, raise exceptions with `raise ... from err` or `raise ... from None` to distinguish them from errors in exception handling
+    "B905",   # `zip()` without an explicit `strict=` parameter
+    "E402",   # Module level import not at top of file
+    "E714",   # Test for object identity should be `is not`
+    "E721",   # Use `is` and `is not` for type comparisons, or `isinstance()` for isinstance checks
+    "E731",   # Do not assign a `lambda` expression, use a `def`
+    "E741",   # Ambiguous variable name: `l`
+    # The following rules will make extensive changes, we will adopt these gradually in the future
+    "UP007",  # Update Union type annotations to use the | operator
+    "UP031",  # Replace printf style string formatting with f-strings
+    "UP032",  # Replace .format() style string formatting with f-strings
+    "UP033",  # Use functools.cache instead of functools.lru_cache
+]
+
+[tool.ruff.lint.extend-per-file-ignores]
+"__init__.py" = ["F401", "F403", "F811"]
+"dask/tokenize.py" = ["F811"]
+"*_test.py" = ["B011"]
+"**/tests/*" = ["B011", "F811"]
+
 [tool.isort]
 sections = ["FUTURE", "STDLIB", "THIRDPARTY", "DISTRIBUTED", "FIRSTPARTY", "LOCALFOLDER"]
 profile = "black"

Original file line number	Diff line number	Diff line change
`@@ -1256,7 +1256,7 @@ def solve(a, b, sym_pos=None, assume_a="gen"):`
`1256`	`1256`	`b = p.T.dot(b)`
`1257`	`1257`	`else:`
`1258`	`1258`	`raise ValueError(`
`1259`		`- f"{assume_a = } is not a recognized matrix structure, " # noqa: E251`
	`1259`	`+ f"{assume_a = } is not a recognized matrix structure, "`
`1260`	`1260`	`"valid structures in Dask are 'pos' and 'gen'."`
`1261`	`1261`	`)`
`1262`	`1262`