chore(python): Fix flake8-pytest-style errors in tests. (#6031)

py-polars/pyproject.toml

@@ -118,6 +118,8 @@ select = [
   "Q",
   # pyupgrade
   "UP",
+  # flake8-pytest-style
+  "PT",
 ]
 ignore = []
 
@@ -134,6 +136,8 @@ extend-ignore = [
   "D402",
   "D415",
   "D416",
+  # flake8-pytest-style:
+  "PT011", # pytest.raises({exception}) is too broad, set the match parameter or use a more specific exception
   # TODO: Remove errors below to further improve docstring linting
   # Ordered from most common to least common errors.
   "D105",

py-polars/requirements-lint.txt

@@ -1,3 +1,3 @@
 black==22.8.0
 blackdoc==0.3.7
-ruff==0.0.208
+ruff==0.0.209

py-polars/tests/slow/conftest.py

@@ -7,6 +7,6 @@
 )
 
 
-@pytest.fixture
+@pytest.fixture()
 def io_test_dir() -> str:
     return IO_TEST_DIR

py-polars/tests/unit/conftest.py

@@ -46,22 +46,22 @@
 )
 
 
-@pytest.fixture
+@pytest.fixture()
 def io_test_dir() -> str:
     return IO_TEST_DIR
 
 
-@pytest.fixture
+@pytest.fixture()
 def examples_dir() -> str:
     return EXAMPLES_DIR
 
 
-@pytest.fixture
+@pytest.fixture()
 def foods_csv() -> str:
     return FOODS_CSV
 
 
-@pytest.fixture
+@pytest.fixture()
 def foods_csv_glob() -> str:
     return FOODS_CSV
 
@@ -76,22 +76,22 @@ def foods_csv_glob() -> str:
     pl.read_csv(FOODS_CSV).write_json(FOODS_NDJSON, json_lines=True)
 
 
-@pytest.fixture
+@pytest.fixture()
 def foods_ipc() -> str:
     return FOODS_IPC
 
 
-@pytest.fixture
+@pytest.fixture()
 def foods_parquet() -> str:
     return FOODS_PARQUET
 
 
-@pytest.fixture
+@pytest.fixture()
 def foods_ndjson() -> str:
     return FOODS_NDJSON
 
 
-@pytest.fixture
+@pytest.fixture()
 def df() -> pl.DataFrame:
     df = pl.DataFrame(
         {
@@ -122,14 +122,14 @@ def df() -> pl.DataFrame:
     )
 
 
-@pytest.fixture
+@pytest.fixture()
 def df_no_lists(df: pl.DataFrame) -> pl.DataFrame:
     return df.select(
         pl.all().exclude(["list_str", "list_int", "list_bool", "list_int", "list_flt"])
     )
 
 
-@pytest.fixture
+@pytest.fixture()
 def fruits_cars() -> pl.DataFrame:
     return pl.DataFrame(
         {

py-polars/tests/unit/io/test_avro.py

@@ -8,12 +8,12 @@
 import polars as pl
 
 
-@pytest.fixture
+@pytest.fixture()
 def example_df() -> pl.DataFrame:
     return pl.DataFrame({"i64": [1, 2], "f64": [0.1, 0.2], "utf8": ["a", "b"]})
 
 
-@pytest.fixture
+@pytest.fixture()
 def compressions() -> list[str]:
     return ["uncompressed", "snappy", "deflate"]
 

py-polars/tests/unit/io/test_csv.py

@@ -232,7 +232,8 @@ def test_partial_column_rename() -> None:
 
 
 @pytest.mark.parametrize(
-    "col_input, col_out", [([0, 1], ["a", "b"]), ([0, 2], ["a", "c"]), (["b"], ["b"])]
+    ("col_input", "col_out"),
+    [([0, 1], ["a", "b"]), ([0, 2], ["a", "c"]), (["b"], ["b"])],
 )
 def test_read_csv_columns_argument(
     col_input: list[int] | list[str], col_out: list[str]
@@ -753,7 +754,7 @@ def test_csv_dtype_overwrite_bool() -> None:
 
 
 @pytest.mark.parametrize(
-    "fmt,expected",
+    ("fmt", "expected"),
     [
         (None, "dt\n2022-01-02T00:00:00.000000\n"),
         ("%F %T%.3f", "dt\n2022-01-02 00:00:00.000\n"),
@@ -770,7 +771,7 @@ def test_datetime_format(fmt: str, expected: str) -> None:
 
 
 @pytest.mark.parametrize(
-    "tu1,tu2,expected",
+    ("tu1", "tu2", "expected"),
     [
         (
             "ns",
@@ -810,7 +811,7 @@ def test_datetime_format_inferred_precision(
 
 
 @pytest.mark.parametrize(
-    "fmt,expected",
+    ("fmt", "expected"),
     [
         (None, "dt\n2022-01-02\n"),
         ("%Y", "dt\n2022\n"),
@@ -825,7 +826,7 @@ def test_date_format(fmt: str, expected: str) -> None:
 
 
 @pytest.mark.parametrize(
-    "fmt,expected",
+    ("fmt", "expected"),
     [
         (None, "dt\n16:15:30.000000000\n"),
         ("%R", "dt\n16:15\n"),

py-polars/tests/unit/io/test_ipc.py

@@ -11,7 +11,7 @@
 from polars.testing import assert_frame_equal_local_categoricals
 
 
-@pytest.fixture
+@pytest.fixture()
 def compressions() -> list[str]:
     return ["uncompressed", "lz4", "zstd"]
 

py-polars/tests/unit/io/test_parquet.py

@@ -26,7 +26,7 @@
 ]
 
 
-@pytest.fixture
+@pytest.fixture()
 def compressions() -> list[ParquetCompression]:
     return COMPRESSIONS
 
@@ -36,17 +36,22 @@ def test_to_from_buffer(
 ) -> None:
     for compression in compressions:
         if compression == "lzo":
-            # lzo compression is not supported now
+            buf = io.BytesIO()
+            # Writing lzo compressed parquet files is not supported for now.
             with pytest.raises(pl.ArrowError):
-                buf = io.BytesIO()
                 df.write_parquet(buf, compression=compression)
-                buf.seek(0)
+            buf.seek(0)
+            # Invalid parquet file as writing failed.
+            with pytest.raises(pl.ArrowError):
                 _ = pl.read_parquet(buf)
 
+            buf = io.BytesIO()
             with pytest.raises(OSError):
-                buf = io.BytesIO()
+                # Writing lzo compressed parquet files is not supported for now.
                 df.write_parquet(buf, compression=compression, use_pyarrow=True)
-                buf.seek(0)
+            buf.seek(0)
+            # Invalid parquet file as writing failed.
+            with pytest.raises(pl.ArrowError):
                 _ = pl.read_parquet(buf)
         else:
             buf = io.BytesIO()
@@ -69,13 +74,18 @@ def test_to_from_file(
     f = os.path.join(io_test_dir, "small.parquet")
     for compression in compressions:
         if compression == "lzo":
-            # lzo compression is not supported now
+            # Writing lzo compressed parquet files is not supported for now.
             with pytest.raises(pl.ArrowError):
                 df.write_parquet(f, compression=compression)
+            # Invalid parquet file as writing failed.
+            with pytest.raises(pl.ArrowError):
                 _ = pl.read_parquet(f)
 
+            # Writing lzo compressed parquet files is not supported for now.
             with pytest.raises(OSError):
                 df.write_parquet(f, compression=compression, use_pyarrow=True)
+            # Invalid parquet file as writing failed.
+            with pytest.raises(FileNotFoundError):
                 _ = pl.read_parquet(f)
         else:
             df.write_parquet(f, compression=compression)

py-polars/tests/unit/test_cfg.py

@@ -9,7 +9,7 @@
 
 
 @pytest.fixture(autouse=True)
-def environ() -> Iterator[None]:
+def _environ() -> Iterator[None]:
     """Fixture to restore the environment variables/state after the test."""
     with pl.StringCache(), pl.Config():
         yield

py-polars/tests/unit/test_constructors.py

@@ -89,7 +89,8 @@ def test_init_dict() -> None:
     assert df.schema == {"c": pl.Int8, "d": pl.Int16}
 
     dfe = df.cleared()
-    assert (df.schema == dfe.schema) and (len(dfe) == 0)
+    assert df.schema == dfe.schema
+    assert len(dfe) == 0
 
 
 def test_init_ndarray(monkeypatch: Any) -> None:
@@ -435,7 +436,8 @@ def test_init_only_columns() -> None:
 
         dfe = df.cleared()
         assert len(dfe) == 0
-        assert (df.schema == dfe.schema) and (dfe.shape == df.shape)
+        assert df.schema == dfe.schema
+        assert dfe.shape == df.shape
 
 
 def test_from_dicts_list_without_dtype() -> None:

py-polars/tests/unit/test_datatypes.py

@@ -67,7 +67,7 @@ def test_dtypes_hashable() -> None:
 
 
 @pytest.mark.parametrize(
-    ["dtype", "representation"],
+    ("dtype", "representation"),
     [
         (pl.Boolean, "Boolean"),
         (pl.Datetime, "Datetime"),

py-polars/tests/unit/test_datelike.py

@@ -627,7 +627,7 @@ def test_date_range_lazy() -> None:
 
 
 @pytest.mark.parametrize(
-    "one,two",
+    ("one", "two"),
     [
         (date(2001, 1, 1), date(2001, 1, 2)),
         (datetime(2001, 1, 1), datetime(2001, 1, 2)),

py-polars/tests/unit/test_df.py

@@ -505,7 +505,7 @@ def test_groupby_agg_input_types(lazy: bool) -> None:
     df_or_lazy: pl.DataFrame | pl.LazyFrame = df.lazy() if lazy else df
 
     for bad_param in bad_agg_parameters():
-        with pytest.raises(TypeError):
+        with pytest.raises(TypeError):  # noqa: PT012
             result = df_or_lazy.groupby("a").agg(bad_param)
             if lazy:
                 result.collect()  # type: ignore[union-attr]
@@ -525,7 +525,7 @@ def test_groupby_rolling_agg_input_types(lazy: bool) -> None:
     df_or_lazy: pl.DataFrame | pl.LazyFrame = df.lazy() if lazy else df
 
     for bad_param in bad_agg_parameters():
-        with pytest.raises(TypeError):
+        with pytest.raises(TypeError):  # noqa: PT012
             result = df_or_lazy.groupby_rolling(
                 index_column="index_column", period="2i"
             ).agg(bad_param)
@@ -549,7 +549,7 @@ def test_groupby_dynamic_agg_input_types(lazy: bool) -> None:
     df_or_lazy: pl.DataFrame | pl.LazyFrame = df.lazy() if lazy else df
 
     for bad_param in bad_agg_parameters():
-        with pytest.raises(TypeError):
+        with pytest.raises(TypeError):  # noqa: PT012
             result = df_or_lazy.groupby_dynamic(
                 index_column="index_column", every="2i", closed="right"
             ).agg(bad_param)
@@ -568,7 +568,7 @@ def test_groupby_dynamic_agg_input_types(lazy: bool) -> None:
 
 
 @pytest.mark.parametrize(
-    "stack,exp_shape,exp_columns",
+    ("stack", "exp_shape", "exp_columns"),
     [
         ([pl.Series("stacked", [-1, -1, -1])], (3, 3), ["a", "b", "stacked"]),
         (
@@ -707,9 +707,11 @@ def test_read_missing_file() -> None:
     with pytest.raises(FileNotFoundError, match="fake_csv_file"):
         pl.read_csv("fake_csv_file")
 
-    with pytest.raises(FileNotFoundError, match="fake_csv_file"):
-        with open("fake_csv_file") as f:
-            pl.read_csv(f)
+    with pytest.raises(FileNotFoundError, match="fake_ipc_file"):
+        pl.read_ipc("fake_ipc_file")
+
+    with pytest.raises(FileNotFoundError, match="fake_avro_file"):
+        pl.read_ipc("fake_avro_file")
 
 
 def test_melt() -> None:
@@ -1987,7 +1989,9 @@ def test_get_item() -> None:
     assert df[4, [5]].frame_equal(pl.DataFrame({"fo5": [1024]}))
 
 
-@pytest.mark.parametrize("as_series,inner_dtype", [(True, pl.Series), (False, list)])
+@pytest.mark.parametrize(
+    ("as_series", "inner_dtype"), [(True, pl.Series), (False, list)]
+)
 def test_to_dict(as_series: bool, inner_dtype: Any) -> None:
     df = pl.DataFrame(
         {
@@ -2512,14 +2516,14 @@ def test_item() -> None:
     df = pl.DataFrame({"a": [1]})
     assert df.item() == 1
 
+    df = pl.DataFrame({"a": [1, 2]})
     with pytest.raises(ValueError):
-        df = pl.DataFrame({"a": [1, 2]})
         df.item()
 
+    df = pl.DataFrame({"a": [1], "b": [2]})
     with pytest.raises(ValueError):
-        df = pl.DataFrame({"a": [1], "b": [2]})
         df.item()
 
+    df = pl.DataFrame({})
     with pytest.raises(ValueError):
-        df = pl.DataFrame({})
         df.item()

py-polars/tests/unit/test_fmt.py

@@ -6,7 +6,7 @@
 
 
 @pytest.mark.parametrize(
-    "expected, values",
+    ("expected", "values"),
     [
         pytest.param(
             """shape: (1,)

py-polars/tests/unit/test_lazy.py

@@ -1597,8 +1597,8 @@ def test_from_epoch(input_dtype: PolarsDataType) -> None:
 
     assert_frame_equal(ldf_result, expected)
 
+    ts_col = pl.col("timestamp_s")
     with pytest.raises(ValueError):
-        ts_col = pl.col("timestamp_s")
         _ = ldf.select(pl.from_epoch(ts_col, unit="s2"))  # type: ignore[call-overload]
 
 

py-polars/tests/unit/test_pivot.py

@@ -48,7 +48,7 @@ def test_pivot_list() -> None:
 
 
 @pytest.mark.parametrize(
-    ["agg_fn", "expected_rows"],
+    ("agg_fn", "expected_rows"),
     [
         ("first", [("a", 2, None, None), ("b", None, None, 10)]),
         ("count", [("a", 2, None, None), ("b", None, 2, 1)]),
@@ -74,7 +74,7 @@ def test_pivot_aggregate(agg_fn: PivotAgg, expected_rows: list[tuple[Any]]) -> N
 
 
 @pytest.mark.parametrize(
-    ["agg_fn", "expected_rows"],
+    ("agg_fn", "expected_rows"),
     [
         ("first", [("a", 2, None, None), ("b", None, None, 10)]),
         ("count", [("a", 2, None, None), ("b", None, 2, 1)]),