chore: Bump pyarrow min version to 13.0.0 (#2825)

Author: dangotbanned

.github/workflows/extremes.yml

@@ -25,7 +25,7 @@ jobs:
           cache-suffix: min-versions-${{ matrix.python-version }}
           cache-dependency-glob: "pyproject.toml"
       - name: install-minimum-versions
-        run: uv pip install pipdeptree tox virtualenv setuptools pandas==1.1.3 polars==0.20.4 numpy==1.19.3 pyarrow==11.0.0 "pyarrow-stubs<17" scipy==1.6.0 scikit-learn==1.1.0 duckdb==1.0 tzdata --system
+        run: uv pip install pipdeptree tox virtualenv setuptools pandas==1.1.3 polars==0.20.4 numpy==1.19.3 pyarrow==13.0.0 "pyarrow-stubs<17" scipy==1.6.0 scikit-learn==1.1.0 duckdb==1.0 tzdata --system
       - name: install-reqs
         run: |
           uv pip install -e . --group tests --system
@@ -37,7 +37,7 @@ jobs:
           echo "$DEPS" | grep 'pandas==1.1.3'
           echo "$DEPS" | grep 'polars==0.20.4'
           echo "$DEPS" | grep 'numpy==1.19.3'
-          echo "$DEPS" | grep 'pyarrow==11.0.0'
+          echo "$DEPS" | grep 'pyarrow==13.0.0'
           echo "$DEPS" | grep 'scipy==1.6.0'
           echo "$DEPS" | grep 'scikit-learn==1.1.0'
           echo "$DEPS" | grep 'duckdb==1.0'
@@ -62,7 +62,7 @@ jobs:
           cache-suffix: pretty-old-versions-${{ matrix.python-version }}
           cache-dependency-glob: "pyproject.toml"
       - name: install-pretty-old-versions
-        run: uv pip install pipdeptree tox virtualenv setuptools pandas==1.1.5 polars==0.20.4 numpy==1.19.3 pyarrow==11.0.0 "pyarrow-stubs<17" scipy==1.6.0 scikit-learn==1.1.0 duckdb==1.0 tzdata --system
+        run: uv pip install pipdeptree tox virtualenv setuptools pandas==1.1.5 polars==0.20.4 numpy==1.19.3 pyarrow==14.0.0 "pyarrow-stubs<17" scipy==1.6.0 scikit-learn==1.1.0 duckdb==1.0 tzdata --system
       - name: install-reqs
         run: uv pip install -e . --group tests --system
       - name: show-deps
@@ -75,7 +75,7 @@ jobs:
           echo "$DEPS" | grep 'pandas==1.1.5'
           echo "$DEPS" | grep 'polars==0.20.4'
           echo "$DEPS" | grep 'numpy==1.19.3'
-          echo "$DEPS" | grep 'pyarrow==11.0.0'
+          echo "$DEPS" | grep 'pyarrow==14.0.0'
           echo "$DEPS" | grep 'scipy==1.6.0'
           echo "$DEPS" | grep 'scikit-learn==1.1.0'
           echo "$DEPS" | grep 'duckdb==1.0'

narwhals/_arrow/group_by.py

@@ -122,12 +122,6 @@ def agg(self, *exprs: ArrowExpr) -> ArrowDataFrame:
         ]
         new_column_names = [new_column_names[i] for i in index_map]
         result_simple = result_simple.rename_columns(new_column_names)
-        if self.compliant._backend_version < (12, 0, 0):
-            columns = result_simple.column_names
-            result_simple = result_simple.select(
-                [*self._keys, *[col for col in columns if col not in self._keys]]
-            )
-
         return self.compliant._with_native(result_simple).rename(
             dict(zip(self._keys, self._output_key_names))
         )

narwhals/_arrow/series.py

@@ -29,7 +29,6 @@
     generate_temporary_column_name,
     is_list_of,
     not_implemented,
-    requires,
 )
 from narwhals.dependencies import is_numpy_array_1d
 from narwhals.exceptions import InvalidOperationError, ShapeError
@@ -156,7 +155,7 @@ def from_iterable(
         )
 
     def _from_scalar(self, value: Any) -> Self:
-        if self._backend_version < (13,) and hasattr(value, "as_py"):
+        if hasattr(value, "as_py"):
             value = value.as_py()
         return super()._from_scalar(value)
 
@@ -866,7 +865,6 @@ def cum_count(self, *, reverse: bool) -> Self:
         dtypes = self._version.dtypes
         return (~self.is_null()).cast(dtypes.UInt32()).cum_sum(reverse=reverse)
 
-    @requires.backend_version((13,))
     def cum_min(self, *, reverse: bool) -> Self:
         result = (
             pc.cumulative_min(self.native, skip_nulls=True)
@@ -875,7 +873,6 @@ def cum_min(self, *, reverse: bool) -> Self:
         )
         return self._with_native(result)
 
-    @requires.backend_version((13,))
     def cum_max(self, *, reverse: bool) -> Self:
         result = (
             pc.cumulative_max(self.native, skip_nulls=True)
@@ -884,7 +881,6 @@ def cum_max(self, *, reverse: bool) -> Self:
         )
         return self._with_native(result)
 
-    @requires.backend_version((13,))
     def cum_prod(self, *, reverse: bool) -> Self:
         result = (
             pc.cumulative_prod(self.native, skip_nulls=True)
@@ -1023,7 +1019,6 @@ def rank(self, method: RankMethod, *, descending: bool) -> Self:
         result = pc.if_else(null_mask, lit(None, rank.type), rank)
         return self._with_native(result)
 
-    @requires.backend_version((13,))
     def hist(  # noqa: C901, PLR0912, PLR0915
         self,
         bins: list[float | int] | None,

narwhals/_arrow/series_str.py

@@ -70,13 +70,9 @@ def zfill(self, width: int) -> ArrowSeries:
         binary_join: Incomplete = pc.binary_join_element_wise
         native = self.native
         hyphen, plus = lit("-"), lit("+")
-
-        _slice_length: int | None = (
-            self.len_chars().max() if self.backend_version < (13, 0) else None
-        )
         first_char, remaining_chars = (
             self.slice(0, 1).native,
-            self.slice(1, _slice_length).native,
+            self.slice(1, None).native,
         )
 
         # Conditions

narwhals/_arrow/utils.py

@@ -295,9 +295,8 @@ def cast_for_truediv(
     if pa.types.is_integer(arrow_array.type) and pa.types.is_integer(pa_object.type):
         # GH: 56645.  # noqa: ERA001
         # https://github.com/apache/arrow/issues/35563
-        # NOTE: `pyarrow==11.*` doesn't allow keywords in `Array.cast`
-        return pc.cast(arrow_array, pa.float64(), safe=False), pc.cast(
-            pa_object, pa.float64(), safe=False
+        return arrow_array.cast(pa.float64(), safe=False), pa_object.cast(
+            pa.float64(), safe=False
         )
 
     return arrow_array, pa_object

narwhals/_pandas_like/utils.py

@@ -508,7 +508,7 @@ def narwhals_to_native_dtype(  # noqa: C901, PLR0912, PLR0915
         try:
             import pyarrow as pa  # ignore-banned-import
         except ModuleNotFoundError:  # pragma: no cover
-            msg = "'pyarrow>=11.0.0' is required for `Date` dtype."
+            msg = "'pyarrow>=13.0.0' is required for `Date` dtype."
         return "date32[pyarrow]"
     if isinstance_or_issubclass(dtype, dtypes.Enum):
         if version is Version.V1:

narwhals/_utils.py

@@ -595,7 +595,7 @@ def _backend_version(self) -> tuple[int, ...]:
     Implementation.PANDAS: (1, 1, 3),
     Implementation.MODIN: (0, 8, 2),
     Implementation.CUDF: (24, 10),
-    Implementation.PYARROW: (11,),
+    Implementation.PYARROW: (13,),
     Implementation.PYSPARK: (3, 5),
     Implementation.PYSPARK_CONNECT: (3, 5),
     Implementation.POLARS: (0, 20, 4),

pyproject.toml

@@ -37,7 +37,7 @@ Repository = "https://github.com/narwhals-dev/narwhals"
 pandas = ["pandas>=1.1.3"]
 modin = ["modin"]
 cudf = ["cudf>=24.10.0"]
-pyarrow = ["pyarrow>=11.0.0"]
+pyarrow = ["pyarrow>=13.0.0"]
 pyspark = ["pyspark>=3.5.0"]
 pyspark-connect = ["pyspark[connect]>=3.5.0"]
 polars = ["polars>=0.20.4"]

tests/expr_and_series/cum_max_test.py

@@ -7,7 +7,6 @@
     DUCKDB_VERSION,
     PANDAS_VERSION,
     POLARS_VERSION,
-    PYARROW_VERSION,
     Constructor,
     ConstructorEager,
     assert_equal_data,
@@ -22,12 +21,7 @@
 def test_cum_max_expr(
     request: pytest.FixtureRequest, constructor_eager: ConstructorEager, *, reverse: bool
 ) -> None:
-    if PYARROW_VERSION < (13, 0, 0) and "pyarrow_table" in str(constructor_eager):
-        request.applymarker(pytest.mark.xfail)
-
-    if (PANDAS_VERSION < (2, 1) or PYARROW_VERSION < (13,)) and "pandas_pyarrow" in str(
-        constructor_eager
-    ):
+    if (PANDAS_VERSION < (2, 1)) and "pandas_pyarrow" in str(constructor_eager):
         request.applymarker(pytest.mark.xfail)
 
     name = "reverse_cum_max" if reverse else "cum_max"
@@ -141,8 +135,6 @@ def test_lazy_cum_max_ungrouped(
         "duckdb" in str(constructor) and DUCKDB_VERSION < (1, 3)
     ):
         pytest.skip(reason="too old version")
-    if PYARROW_VERSION < (13, 0, 0) and "pyarrow_table" in str(constructor):
-        request.applymarker(pytest.mark.xfail)
 
     df = nw.from_native(
         constructor({"a": [2, 3, 1], "b": [0, 2, 1], "i": [1, 2, 0]})
@@ -174,8 +166,6 @@ def test_lazy_cum_max_ungrouped_ordered_by_nulls(
         "duckdb" in str(constructor) and DUCKDB_VERSION < (1, 3)
     ):
         pytest.skip(reason="too old version")
-    if PYARROW_VERSION < (13, 0, 0) and "pyarrow_table" in str(constructor):
-        request.applymarker(pytest.mark.xfail)
 
     df = nw.from_native(
         constructor(
@@ -200,12 +190,7 @@ def test_lazy_cum_max_ungrouped_ordered_by_nulls(
 def test_cum_max_series(
     request: pytest.FixtureRequest, constructor_eager: ConstructorEager
 ) -> None:
-    if PYARROW_VERSION < (13, 0, 0) and "pyarrow_table" in str(constructor_eager):
-        request.applymarker(pytest.mark.xfail)
-
-    if (PANDAS_VERSION < (2, 1) or PYARROW_VERSION < (13,)) and "pandas_pyarrow" in str(
-        constructor_eager
-    ):
+    if (PANDAS_VERSION < (2, 1)) and "pandas_pyarrow" in str(constructor_eager):
         request.applymarker(pytest.mark.xfail)
 
     df = nw.from_native(constructor_eager(data), eager_only=True)

tests/expr_and_series/cum_min_test.py

@@ -7,7 +7,6 @@
     DUCKDB_VERSION,
     PANDAS_VERSION,
     POLARS_VERSION,
-    PYARROW_VERSION,
     Constructor,
     ConstructorEager,
     assert_equal_data,
@@ -22,12 +21,7 @@
 def test_cum_min_expr(
     request: pytest.FixtureRequest, constructor_eager: ConstructorEager, *, reverse: bool
 ) -> None:
-    if PYARROW_VERSION < (13, 0, 0) and "pyarrow_table" in str(constructor_eager):
-        request.applymarker(pytest.mark.xfail)
-
-    if (PANDAS_VERSION < (2, 1) or PYARROW_VERSION < (13,)) and "pandas_pyarrow" in str(
-        constructor_eager
-    ):
+    if (PANDAS_VERSION < (2, 1)) and "pandas_pyarrow" in str(constructor_eager):
         request.applymarker(pytest.mark.xfail)
 
     name = "reverse_cum_min" if reverse else "cum_min"
@@ -141,9 +135,6 @@ def test_lazy_cum_min_ungrouped(
         "duckdb" in str(constructor) and DUCKDB_VERSION < (1, 3)
     ):
         pytest.skip(reason="too old version")
-    if PYARROW_VERSION < (13, 0, 0) and "pyarrow_table" in str(constructor):
-        request.applymarker(pytest.mark.xfail)
-
     df = nw.from_native(
         constructor({"a": [2, 3, 1], "b": [0, 2, 1], "i": [1, 2, 0]})
     ).sort("i")
@@ -174,9 +165,6 @@ def test_lazy_cum_min_ungrouped_ordered_by_nulls(
         "duckdb" in str(constructor) and DUCKDB_VERSION < (1, 3)
     ):
         pytest.skip(reason="too old version")
-    if PYARROW_VERSION < (13, 0, 0) and "pyarrow_table" in str(constructor):
-        request.applymarker(pytest.mark.xfail)
-
     df = nw.from_native(
         constructor(
             {
@@ -200,12 +188,7 @@ def test_lazy_cum_min_ungrouped_ordered_by_nulls(
 def test_cum_min_series(
     request: pytest.FixtureRequest, constructor_eager: ConstructorEager
 ) -> None:
-    if PYARROW_VERSION < (13, 0, 0) and "pyarrow_table" in str(constructor_eager):
-        request.applymarker(pytest.mark.xfail)
-
-    if (PANDAS_VERSION < (2, 1) or PYARROW_VERSION < (13,)) and "pandas_pyarrow" in str(
-        constructor_eager
-    ):
+    if (PANDAS_VERSION < (2, 1)) and "pandas_pyarrow" in str(constructor_eager):
         request.applymarker(pytest.mark.xfail)
 
     df = nw.from_native(constructor_eager(data), eager_only=True)