ci: update pre-commit hooks and fix linting issues

* Update Ruff version in pre-commit configuration to v0.15.1.
* Add noqa comments to suppress specific linting warnings in various files.
* Update regex patterns in test cases for better matching.

Author: dariocurr

.pre-commit-config.yaml

@@ -22,7 +22,7 @@ repos:
           - id: actionlint-docker
       - repo: https://github.com/astral-sh/ruff-pre-commit
         # Ruff version.
-        rev: v0.9.10
+        rev: v0.15.1
         hooks:
           # Run the linter.
           - id: ruff

python/datafusion/dataframe.py

@@ -330,7 +330,7 @@ def into_view(self, temporary: bool = False) -> Table:
             >>> df.collect()  # The DataFrame is still usable
             >>> ctx.sql("SELECT value FROM values_view").collect()
         """
-        from datafusion.catalog import Table as _Table
+        from datafusion.catalog import Table as _Table  # noqa: PLC0415
 
         return _Table(self.df.into_view(temporary))
 

python/datafusion/expr.py

@@ -20,6 +20,8 @@
 See :ref:`Expressions` in the online documentation for more details.
 """
 
+# ruff: noqa: PLC0415
+
 from __future__ import annotations
 
 from collections.abc import Iterable, Sequence
@@ -340,7 +342,7 @@ def sort_list_to_raw_sort_list(
     return raw_sort_list
 
 
-class Expr:
+class Expr:  # noqa: PLW1641
     """Expression object.
 
     Expressions are one of the core concepts in DataFusion. See

python/datafusion/input/location.py

@@ -46,7 +46,7 @@ def build_table(
         num_rows = 0  # Total number of rows in the file. Used for statistics
         columns = []
         if file_format == "parquet":
-            import pyarrow.parquet as pq
+            import pyarrow.parquet as pq  # noqa: PLC0415
 
             # Read the Parquet metadata
             metadata = pq.read_metadata(input_item)
@@ -61,7 +61,7 @@ def build_table(
             ]
 
         elif format == "csv":
-            import csv
+            import csv  # noqa: PLC0415
 
             # Consume header row and count number of rows for statistics.
             # TODO: Possibly makes sense to have the eager number of rows

python/datafusion/plan.py

@@ -32,7 +32,7 @@
 ]
 
 
-class LogicalPlan:
+class LogicalPlan:  # noqa: PLW1641
     """Logical Plan.
 
     A `LogicalPlan` is a node in a tree of relational operators (such as

python/datafusion/user_defined.py

@@ -583,11 +583,11 @@ def from_pycapsule(func: AggregateUDFExportable | _PyCapsule) -> AggregateUDF:
         AggregateUDF that is exported via the FFI bindings.
         """
         if _is_pycapsule(func):
-            aggregate = cast(AggregateUDF, object.__new__(AggregateUDF))
+            aggregate = cast("AggregateUDF", object.__new__(AggregateUDF))
             aggregate._udaf = df_internal.AggregateUDF.from_pycapsule(func)
             return aggregate
 
-        capsule = cast(AggregateUDFExportable, func)
+        capsule = cast("AggregateUDFExportable", func)
         name = str(capsule.__class__)
         return AggregateUDF(
             name=name,

python/tests/test_catalog.py

@@ -248,7 +248,7 @@ def test_exception_not_mangled(ctx: SessionContext):
 
     schema.register_table("test_table", create_dataset())
 
-    with pytest.raises(ValueError, match="^test_table is not an acceptable name$"):
+    with pytest.raises(ValueError, match=r"^test_table is not an acceptable name$"):
         ctx.sql(f"select * from {catalog_name}.{schema_name}.test_table")
 
 

python/tests/test_dataframe.py

@@ -2790,7 +2790,7 @@ def test_write_parquet_with_options_encoding(tmp_path, encoding, data_types, res
 def test_write_parquet_with_options_unsupported_encoding(df, tmp_path, encoding):
     """Test that unsupported Parquet encodings do not work."""
     # BaseException is used since this throws a Rust panic: https://github.com/PyO3/pyo3/issues/3519
-    with pytest.raises(BaseException, match="Encoding .*? is not supported"):
+    with pytest.raises(BaseException, match=r"Encoding .*? is not supported"):
         df.write_parquet_with_options(tmp_path, ParquetWriterOptions(encoding=encoding))
 
 

python/tests/test_functions.py

@@ -303,31 +303,31 @@ def py_flatten(arr):
             lambda data: [np.concatenate([arr, arr]) for arr in data],
         ),
         (
-            lambda col: f.array_dims(col),
+            f.array_dims,
             lambda data: [[len(r)] for r in data],
         ),
         (
-            lambda col: f.array_distinct(col),
+            f.array_distinct,
             lambda data: [list(set(r)) for r in data],
         ),
         (
-            lambda col: f.list_distinct(col),
+            f.list_distinct,
             lambda data: [list(set(r)) for r in data],
         ),
         (
-            lambda col: f.list_dims(col),
+            f.list_dims,
             lambda data: [[len(r)] for r in data],
         ),
         (
             lambda col: f.array_element(col, literal(1)),
             lambda data: [r[0] for r in data],
         ),
         (
-            lambda col: f.array_empty(col),
+            f.array_empty,
             lambda data: [len(r) == 0 for r in data],
         ),
         (
-            lambda col: f.empty(col),
+            f.empty,
             lambda data: [len(r) == 0 for r in data],
         ),
         (
@@ -343,11 +343,11 @@ def py_flatten(arr):
             lambda data: [r[0] for r in data],
         ),
         (
-            lambda col: f.array_length(col),
+            f.array_length,
             lambda data: [len(r) for r in data],
         ),
         (
-            lambda col: f.list_length(col),
+            f.list_length,
             lambda data: [len(r) for r in data],
         ),
         (
@@ -391,11 +391,11 @@ def py_flatten(arr):
             lambda data: [[i + 1 for i, _v in enumerate(r) if _v == 1.0] for r in data],
         ),
         (
-            lambda col: f.array_ndims(col),
+            f.array_ndims,
             lambda data: [np.array(r).ndim for r in data],
         ),
         (
-            lambda col: f.list_ndims(col),
+            f.list_ndims,
             lambda data: [np.array(r).ndim for r in data],
         ),
         (
@@ -415,11 +415,11 @@ def py_flatten(arr):
             lambda data: [np.insert(arr, 0, 99.0) for arr in data],
         ),
         (
-            lambda col: f.array_pop_back(col),
+            f.array_pop_back,
             lambda data: [arr[:-1] for arr in data],
         ),
         (
-            lambda col: f.array_pop_front(col),
+            f.array_pop_front,
             lambda data: [arr[1:] for arr in data],
         ),
         (

python/tests/test_sql.py

@@ -31,7 +31,7 @@
 def test_no_table(ctx):
     with pytest.raises(
         ValueError,
-        match="^Error during planning: table 'datafusion.public.b' not found$",
+        match=r"^Error during planning: table 'datafusion.public.b' not found$",
     ):
         ctx.sql("SELECT a FROM b").collect()
 
@@ -188,7 +188,7 @@ def test_register_parquet_partitioned(ctx, tmp_path, path_to_str, legacy_data_ty
     partition_data_type = "string" if legacy_data_type else pa.string()
 
     if legacy_data_type:
-        with pytest.warns(DeprecationWarning):
+        with pytest.warns(DeprecationWarning):  # noqa: PT030
             ctx.register_parquet(
                 "datapp",
                 dir_root,