test: enhance xfail markers with specific reasons in timestamp_test.py (#2065)

Author: luke396

tests/expr_and_series/dt/timestamp_test.py

@@ -17,7 +17,7 @@
 from tests.utils import Constructor
 from tests.utils import ConstructorEager
 from tests.utils import assert_equal_data
-from tests.utils import is_windows
+from tests.utils import is_pyarrow_windows_no_tzdata
 
 if TYPE_CHECKING:
     from narwhals.typing import IntoSeriesT
@@ -55,15 +55,17 @@ def test_timestamp_datetimes(
     expected: list[int | None],
 ) -> None:
     if any(x in str(constructor) for x in ("duckdb", "pyspark")):
-        request.applymarker(pytest.mark.xfail)
+        request.applymarker(
+            pytest.mark.xfail(reason="Backend timestamp conversion not yet implemented")
+        )
     if original_time_unit == "s" and "polars" in str(constructor):
-        request.applymarker(pytest.mark.xfail)
+        pytest.skip("Second precision not supported in Polars")
+
     if "pandas_pyarrow" in str(constructor) and PANDAS_VERSION < (
         2,
         2,
     ):  # pragma: no cover
-        # pyarrow-backed timestamps were too inconsistent and unreliable before 2.2
-        request.applymarker(pytest.mark.xfail(strict=False))
+        pytest.skip("Requires pandas >= 2.2 for reliable pyarrow-backed timestamps")
     datetimes = {"a": [datetime(2001, 1, 1), None, datetime(2001, 1, 3)]}
     df = nw.from_native(constructor(datetimes))
     result = df.select(
@@ -97,27 +99,38 @@ def test_timestamp_datetimes_tz_aware(
     expected: list[int | None],
 ) -> None:
     if any(x in str(constructor) for x in ("duckdb", "pyspark")):
-        request.applymarker(pytest.mark.xfail)
-    if (
-        (any(x in str(constructor) for x in ("pyarrow",)) and is_windows())
-        or ("pandas_pyarrow" in str(constructor) and PANDAS_VERSION < (2,))
-        or ("pyarrow_table" in str(constructor) and PYARROW_VERSION < (12,))
-    ):
-        request.applymarker(pytest.mark.xfail)
-    if "pandas_pyarrow" in str(constructor) and PANDAS_VERSION < (
-        2,
-        2,
-    ):  # pragma: no cover
-        # pyarrow-backed timestamps were too inconsistent and unreliable before 2.2
-        request.applymarker(pytest.mark.xfail(strict=False))
-    if "dask" in str(constructor) and PANDAS_VERSION < (
-        2,
-        1,
-    ):  # pragma: no cover
-        request.applymarker(pytest.mark.xfail)
+        request.applymarker(
+            pytest.mark.xfail(reason="Backend timestamp conversion not yet implemented")
+        )
+    version_conditions = [
+        (
+            is_pyarrow_windows_no_tzdata(constructor),
+            "Timezone database is not installed on Windows",
+        ),
+        (
+            "pandas_pyarrow" in str(constructor) and PANDAS_VERSION < (2,),
+            "Requires pandas >= 2.0 for pyarrow support",
+        ),
+        (
+            "pyarrow_table" in str(constructor) and PYARROW_VERSION < (12,),
+            "Requires pyarrow >= 12.0",
+        ),
+        (
+            "pandas_pyarrow" in str(constructor) and PANDAS_VERSION < (2, 2),
+            "Requires pandas >= 2.2 for reliable timestamps",
+        ),
+        (
+            "dask" in str(constructor) and PANDAS_VERSION < (2, 1),
+            "Requires pandas >= 2.1 for dask support",
+        ),
+    ]
+
+    for condition, reason in version_conditions:
+        if condition:
+            pytest.skip(reason)  # pragma: no cover
 
     if original_time_unit == "s" and "polars" in str(constructor):
-        request.applymarker(pytest.mark.xfail)
+        request.applymarker(pytest.mark.xfail(reason="Second precision not supported"))
     datetimes = {"a": [datetime(2001, 1, 1), None, datetime(2001, 1, 3)]}
     df = nw.from_native(constructor(datetimes))
     result = df.select(
@@ -145,17 +158,17 @@ def test_timestamp_dates(
     expected: list[int | None],
 ) -> None:
     if any(x in str(constructor) for x in ("duckdb", "pyspark")):
-        request.applymarker(pytest.mark.xfail)
-    if any(
-        x in str(constructor)
-        for x in (
-            "pandas_constructor",
-            "pandas_nullable_constructor",
-            "cudf",
-            "modin_constructor",
+        request.applymarker(
+            pytest.mark.xfail(reason="Backend timestamp conversion not yet implemented")
         )
-    ):
-        request.applymarker(pytest.mark.xfail)
+    unsupported_backends = (
+        "pandas_constructor",
+        "pandas_nullable_constructor",
+        "cudf",
+        "modin_constructor",
+    )
+    if any(x in str(constructor) for x in unsupported_backends):
+        pytest.skip("Backend does not support date type")
 
     dates = {"a": [datetime(2001, 1, 1), None, datetime(2001, 1, 3)]}
     if "dask" in str(constructor):
@@ -172,9 +185,15 @@ def test_timestamp_invalid_date(
     request: pytest.FixtureRequest, constructor: Constructor
 ) -> None:
     if any(x in str(constructor) for x in ("duckdb", "pyspark")):
-        request.applymarker(pytest.mark.xfail)
+        request.applymarker(
+            pytest.mark.xfail(reason="Backend timestamp conversion not yet implemented")
+        )
     if "polars" in str(constructor):
-        request.applymarker(pytest.mark.xfail)
+        request.applymarker(
+            pytest.mark.xfail(
+                reason="Invalid date handling not yet implemented in Polars"
+            )
+        )
     data_str = {"a": ["x", "y", None]}
     data_num = {"a": [1, 2, None]}
     df_str = nw.from_native(constructor(data_str))

tests/utils.py

@@ -4,6 +4,7 @@
 import os
 import sys
 import warnings
+from pathlib import Path
 from typing import Any
 from typing import Callable
 from typing import Iterator
@@ -159,3 +160,13 @@ def maybe_get_modin_df(df_pandas: pd.DataFrame) -> Any:
 def is_windows() -> bool:
     """Check if the current platform is Windows."""
     return sys.platform in {"win32", "cygwin"}
+
+
+def windows_has_tzdata() -> bool:  # pragma: no cover
+    """From PyArrow: python/pyarrow/tests/util.py."""
+    return (Path.home() / "Downloads" / "tzdata").exists()
+
+
+def is_pyarrow_windows_no_tzdata(constructor: Constructor, /) -> bool:
+    """Skip test on Windows when the tz database is not configured."""
+    return "pyarrow" in str(constructor) and is_windows() and not windows_has_tzdata()