drive-by typing (#882)

narwhals/_arrow/dataframe.py

@@ -21,6 +21,7 @@
 
 if TYPE_CHECKING:
     import numpy as np
+    import pyarrow as pa
     from typing_extensions import Self
 
     from narwhals._arrow.group_by import ArrowGroupBy
@@ -33,7 +34,7 @@
 class ArrowDataFrame:
     # --- not in the spec ---
     def __init__(
-        self, native_dataframe: Any, *, backend_version: tuple[int, ...]
+        self, native_dataframe: pa.Table, *, backend_version: tuple[int, ...]
     ) -> None:
         self._native_frame = native_dataframe
         self._implementation = Implementation.PYARROW

narwhals/_arrow/expr.py

@@ -15,6 +15,7 @@
     from narwhals._arrow.dataframe import ArrowDataFrame
     from narwhals._arrow.namespace import ArrowNamespace
     from narwhals._arrow.series import ArrowSeries
+    from narwhals._arrow.typing import IntoArrowExpr
     from narwhals.dtypes import DType
 
 
@@ -157,7 +158,7 @@ def __invert__(self) -> Self:
     def len(self) -> Self:
         return reuse_series_implementation(self, "len", returns_scalar=True)
 
-    def filter(self, *predicates: Any) -> Self:
+    def filter(self, *predicates: IntoArrowExpr) -> Self:
         plx = self.__narwhals_namespace__()
         expr = plx.all_horizontal(*predicates)
         return reuse_series_implementation(self, "filter", other=expr)
@@ -228,7 +229,7 @@ def null_count(self) -> Self:
     def is_null(self) -> Self:
         return reuse_series_implementation(self, "is_null")
 
-    def is_between(self, lower_bound: Any, upper_bound: Any, closed: str) -> Any:
+    def is_between(self, lower_bound: Any, upper_bound: Any, closed: str) -> Self:
         return reuse_series_implementation(
             self, "is_between", lower_bound, upper_bound, closed
         )

narwhals/_arrow/series.py

@@ -18,6 +18,7 @@
 from narwhals.utils import generate_unique_token
 
 if TYPE_CHECKING:
+    import pyarrow as pa
     from typing_extensions import Self
 
     from narwhals._arrow.dataframe import ArrowDataFrame
@@ -26,7 +27,11 @@
 
 class ArrowSeries:
     def __init__(
-        self, native_series: Any, *, name: str, backend_version: tuple[int, ...]
+        self,
+        native_series: pa.ChunkedArray,
+        *,
+        name: str,
+        backend_version: tuple[int, ...],
     ) -> None:
         self._name = name
         self._native_series = native_series
@@ -366,7 +371,9 @@ def all(self) -> bool:
 
         return pc.all(self._native_series)  # type: ignore[no-any-return]
 
-    def is_between(self, lower_bound: Any, upper_bound: Any, closed: str = "both") -> Any:
+    def is_between(
+        self, lower_bound: Any, upper_bound: Any, closed: str = "both"
+    ) -> Self:
         import pyarrow.compute as pc  # ignore-banned-import()
 
         ser = self._native_series
@@ -657,7 +664,7 @@ def clip(
 
         return self._from_native_series(arr)
 
-    def to_arrow(self: Self) -> Any:
+    def to_arrow(self: Self) -> pa.Array:
         return self._native_series.combine_chunks()
 
     @property

pyproject.toml

@@ -24,6 +24,11 @@ exclude = [
   "/tests",
   "/tpch",
   "/utils",
+  ".gitignore",
+  "CONTRIBUTING.md",
+  "mkdocs.yml",
+  "noxfile.py",
+  "requirements-dev.txt",
 ]
 
 [project.optional-dependencies]