test: mark failing dask tests, add dask to constructor (narwhals-dev#697

)

tests/conftest.py

@@ -88,8 +88,8 @@ def pyarrow_table_constructor(obj: Any) -> IntoDataFrame:
 if get_modin() is not None:  # pragma: no cover
     eager_constructors.append(modin_constructor)
 # TODO(unassigned): when Dask gets better support, remove the "False and" part
-if False and get_dask_dataframe() is not None:  # pragma: no cover  # noqa: SIM223
-    lazy_constructors.append(dask_lazy_constructor)
+if get_dask_dataframe() is not None:  # pragma: no cover
+    lazy_constructors.append(dask_lazy_constructor)  # type: ignore  # noqa: PGH003
 
 
 @pytest.fixture(params=eager_constructors)

tests/expr_and_series/abs_test.py

@@ -1,10 +1,14 @@
 from typing import Any
 
+import pytest
+
 import narwhals.stable.v1 as nw
 from tests.utils import compare_dicts
 
 
-def test_abs(constructor: Any) -> None:
+def test_abs(constructor: Any, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     df = nw.from_native(constructor({"a": [1, 2, 3, -4, 5]}))
     result = df.select(b=nw.col("a").abs())
     expected = {"b": [1, 2, 3, 4, 5]}

tests/expr_and_series/any_all_test.py

@@ -1,10 +1,14 @@
 from typing import Any
 
+import pytest
+
 import narwhals.stable.v1 as nw
 from tests.utils import compare_dicts
 
 
-def test_any_all(constructor: Any) -> None:
+def test_any_all(constructor: Any, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     df = nw.from_native(
         constructor(
             {

tests/expr_and_series/any_horizontal_test.py

@@ -1,10 +1,14 @@
 from typing import Any
 
+import pytest
+
 import narwhals.stable.v1 as nw
 from tests.utils import compare_dicts
 
 
-def test_anyh(constructor: Any) -> None:
+def test_anyh(constructor: Any, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     data = {
         "a": [False, False, True],
         "b": [False, True, True],

tests/expr_and_series/arg_true_test.py

@@ -1,10 +1,14 @@
 from typing import Any
 
+import pytest
+
 import narwhals as nw
 from tests.utils import compare_dicts
 
 
-def test_arg_true(constructor: Any) -> None:
+def test_arg_true(constructor: Any, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     df = nw.from_native(constructor({"a": [1, None, None, 3]}))
     result = df.select(nw.col("a").is_null().arg_true())
     expected = {"a": [1, 2]}

tests/expr_and_series/arithmetic_test.py

@@ -22,8 +22,18 @@
     ],
 )
 def test_arithmetic(
-    attr: str, rhs: Any, expected: list[Any], constructor: Any, request: Any
+    attr: str,
+    rhs: Any,
+    expected: list[Any],
+    constructor: Any,
+    request: Any,
 ) -> None:
+    if "dask" in str(constructor) and attr not in [
+        "__add__",
+        "__sub__",
+        "__mul__",
+    ]:
+        request.applymarker(pytest.mark.xfail)
     if attr == "__mod__" and any(
         x in str(constructor) for x in ["pandas_pyarrow", "pyarrow_table", "modin"]
     ):
@@ -48,8 +58,14 @@ def test_arithmetic(
     ],
 )
 def test_right_arithmetic(
-    attr: str, rhs: Any, expected: list[Any], constructor: Any, request: Any
+    attr: str,
+    rhs: Any,
+    expected: list[Any],
+    constructor: Any,
+    request: Any,
 ) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     if attr == "__rmod__" and any(
         x in str(constructor) for x in ["pandas_pyarrow", "pyarrow_table", "modin"]
     ):
@@ -75,7 +91,11 @@ def test_right_arithmetic(
     ],
 )
 def test_arithmetic_series(
-    attr: str, rhs: Any, expected: list[Any], constructor_eager: Any, request: Any
+    attr: str,
+    rhs: Any,
+    expected: list[Any],
+    constructor_eager: Any,
+    request: Any,
 ) -> None:
     if attr == "__mod__" and any(
         x in str(constructor_eager) for x in ["pandas_pyarrow", "pyarrow_table", "modin"]
@@ -101,7 +121,11 @@ def test_arithmetic_series(
     ],
 )
 def test_right_arithmetic_series(
-    attr: str, rhs: Any, expected: list[Any], constructor_eager: Any, request: Any
+    attr: str,
+    rhs: Any,
+    expected: list[Any],
+    constructor_eager: Any,
+    request: Any,
 ) -> None:
     if attr == "__rmod__" and any(
         x in str(constructor_eager) for x in ["pandas_pyarrow", "pyarrow_table", "modin"]

tests/expr_and_series/cast_test.py

@@ -46,6 +46,8 @@
 
 @pytest.mark.filterwarnings("ignore:casting period[M] values to int64:FutureWarning")
 def test_cast(constructor: Any, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     if "pyarrow_table_constructor" in str(constructor) and parse_version(
         pa.__version__
     ) <= (15,):  # pragma: no cover

tests/expr_and_series/count_test.py

@@ -1,10 +1,14 @@
 from typing import Any
 
+import pytest
+
 import narwhals.stable.v1 as nw
 from tests.utils import compare_dicts
 
 
-def test_count(constructor: Any) -> None:
+def test_count(constructor: Any, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     data = {"a": [1, 3, 2], "b": [4, None, 6], "z": [7.0, None, None]}
     df = nw.from_native(constructor(data))
     result = df.select(nw.col("a", "b", "z").count())

tests/expr_and_series/cum_sum_test.py

@@ -1,5 +1,7 @@
 from typing import Any
 
+import pytest
+
 import narwhals.stable.v1 as nw
 from tests.utils import compare_dicts
 
@@ -10,7 +12,9 @@
 }
 
 
-def test_cum_sum_simple(constructor: Any) -> None:
+def test_cum_sum_simple(constructor: Any, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     df = nw.from_native(constructor(data))
     result = df.select(nw.all().cum_sum())
     expected = {

tests/expr_and_series/diff_test.py

@@ -15,6 +15,8 @@
 
 
 def test_diff(constructor: Any, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     if "pyarrow_table_constructor" in str(constructor) and parse_version(
         pa.__version__
     ) < (13,):

tests/expr_and_series/double_selected_test.py

@@ -1,10 +1,14 @@
 from typing import Any
 
+import pytest
+
 import narwhals.stable.v1 as nw
 from tests.utils import compare_dicts
 
 
-def test_double_selected(constructor: Any) -> None:
+def test_double_selected(constructor: Any, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     data = {"a": [1, 3, 2], "b": [4, 4, 6], "z": [7.0, 8, 9]}
     df = nw.from_native(constructor(data))
 

tests/expr_and_series/double_test.py

@@ -1,20 +1,31 @@
 from typing import Any
 
+import pytest
+
 import narwhals.stable.v1 as nw
 from tests.utils import compare_dicts
 
 
-def test_double(constructor: Any) -> None:
+def test_double(constructor: Any, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     data = {"a": [1, 3, 2], "b": [4, 4, 6], "z": [7.0, 8, 9]}
     df = nw.from_native(constructor(data))
     result = df.with_columns(nw.all() * 2)
     expected = {"a": [2, 6, 4], "b": [8, 8, 12], "z": [14.0, 16.0, 18.0]}
     compare_dicts(result, expected)
 
 
-def test_double_alias(constructor: Any) -> None:
+def test_double_alias(constructor: Any, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     data = {"a": [1, 3, 2], "b": [4, 4, 6], "z": [7.0, 8, 9]}
     df = nw.from_native(constructor(data))
     result = df.with_columns(nw.col("a").alias("o"), nw.all() * 2)
-    expected = {"o": [1, 3, 2], "a": [2, 6, 4], "b": [8, 8, 12], "z": [14.0, 16.0, 18.0]}
+    expected = {
+        "o": [1, 3, 2],
+        "a": [2, 6, 4],
+        "b": [8, 8, 12],
+        "z": [14.0, 16.0, 18.0],
+    }
     compare_dicts(result, expected)

tests/expr_and_series/drop_nulls_test.py

@@ -2,11 +2,15 @@
 
 from typing import Any
 
+import pytest
+
 import narwhals as nw
 from tests.utils import compare_dicts
 
 
-def test_drop_nulls(constructor: Any) -> None:
+def test_drop_nulls(constructor: Any, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     data = {
         "A": [1, 2, None, 4],
         "B": [5, 6, 7, 8],

tests/expr_and_series/filter_test.py

@@ -1,5 +1,7 @@
 from typing import Any
 
+import pytest
+
 import narwhals.stable.v1 as nw
 from tests.utils import compare_dicts
 
@@ -11,7 +13,9 @@
 }
 
 
-def test_filter(constructor: Any) -> None:
+def test_filter(constructor: Any, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     df = nw.from_native(constructor(data))
     result = df.select(nw.col("a").filter(nw.col("i") < 2, nw.col("c") == 5))
     expected = {"a": [0]}

tests/expr_and_series/gather_every_test.py

@@ -10,7 +10,9 @@
 
 @pytest.mark.parametrize("n", [1, 2, 3])
 @pytest.mark.parametrize("offset", [1, 2, 3])
-def test_gather_every_expr(constructor: Any, n: int, offset: int) -> None:
+def test_gather_every_expr(constructor: Any, n: int, offset: int, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     df = nw.from_native(constructor(data))
 
     result = df.select(nw.col("a").gather_every(n=n, offset=offset))

tests/expr_and_series/is_between_test.py

@@ -21,7 +21,11 @@
         ("none", [False, True, True, False]),
     ],
 )
-def test_is_between(constructor: Any, closed: str, expected: list[bool]) -> None:
+def test_is_between(
+    constructor: Any, closed: str, expected: list[bool], request: Any
+) -> None:
+    if "dask" in str(constructor) and closed == "none":
+        request.applymarker(pytest.mark.xfail)
     df = nw.from_native(constructor(data))
     result = df.select(nw.col("a").is_between(1, 5, closed=closed))
     expected_dict = {"a": expected}

tests/expr_and_series/is_duplicated_test.py

@@ -13,6 +13,8 @@
 
 
 def test_is_duplicated_expr(constructor: Any, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     if "modin" in str(constructor):
         # TODO(unassigned): why is Modin failing here?
         request.applymarker(pytest.mark.xfail)

tests/expr_and_series/is_first_distinct_test.py

@@ -13,6 +13,8 @@
 
 
 def test_is_first_distinct_expr(constructor: Any, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     if "modin" in str(constructor):
         # TODO(unassigned): why is Modin failing here?
         request.applymarker(pytest.mark.xfail)

tests/expr_and_series/is_in_test.py

@@ -1,5 +1,7 @@
 from typing import Any
 
+import pytest
+
 import narwhals.stable.v1 as nw
 from tests.utils import compare_dicts
 
@@ -9,7 +11,9 @@
 }
 
 
-def test_expr_is_in(constructor: Any) -> None:
+def test_expr_is_in(constructor: Any, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     df = nw.from_native(constructor(data))
     result = df.select(nw.col("a").is_in([4, 5]))
     expected = {"a": [False, True, False, True]}

tests/expr_and_series/is_last_distinct_test.py

@@ -13,6 +13,8 @@
 
 
 def test_is_last_distinct_expr(constructor: Any, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     if "modin" in str(constructor):
         # TODO(unassigned): why is Modin failing here?
         request.applymarker(pytest.mark.xfail)

tests/expr_and_series/is_unique_test.py

@@ -13,6 +13,8 @@
 
 
 def test_is_unique_expr(constructor: Any, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     if "modin" in str(constructor):
         # TODO(unassigned): why is Modin failing here?
         request.applymarker(pytest.mark.xfail)

tests/expr_and_series/len_test.py

@@ -1,13 +1,17 @@
 from typing import Any
 
+import pytest
+
 import narwhals.stable.v1 as nw
 from tests.utils import compare_dicts
 
 data = {"a": list("xyz"), "b": [1, 2, 1]}
 expected = {"a1": [2], "a2": [1]}
 
 
-def test_len(constructor: Any) -> None:
+def test_len(constructor: Any, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     df_raw = constructor(data)
     df = nw.from_native(df_raw).select(
         nw.col("a").filter(nw.col("b") == 1).len().alias("a1"),

tests/expr_and_series/max_test.py

@@ -11,7 +11,9 @@
 
 
 @pytest.mark.parametrize("expr", [nw.col("a", "b", "z").max(), nw.max("a", "b", "z")])
-def test_expr_max_expr(constructor: Any, expr: nw.Expr) -> None:
+def test_expr_max_expr(constructor: Any, expr: nw.Expr, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     df = nw.from_native(constructor(data))
     result = df.select(expr)
     expected = {"a": [3], "b": [6], "z": [9.0]}

tests/expr_and_series/mean_test.py

@@ -11,7 +11,9 @@
 
 
 @pytest.mark.parametrize("expr", [nw.col("a", "b", "z").mean(), nw.mean("a", "b", "z")])
-def test_expr_mean_expr(constructor: Any, expr: nw.Expr) -> None:
+def test_expr_mean_expr(constructor: Any, expr: nw.Expr, request: Any) -> None:
+    if "dask" in str(constructor):
+        request.applymarker(pytest.mark.xfail)
     df = nw.from_native(constructor(data))
     result = df.select(expr)
     expected = {"a": [2.0], "b": [5.0], "z": [8.0]}