TYP: Misc in Algos(#43977)

Author: jbrockmendel

pandas/_libs/algos.pyi

@@ -120,13 +120,6 @@ cdef inline uint64_t _rotl(uint64_t x, uint64_t b) nogil:
     return (x << b) | (x >> (64 - b))
 
 
-cdef inline void u32to8_le(uint8_t* p, uint32_t v) nogil:
-    p[0] = <uint8_t>(v)
-    p[1] = <uint8_t>(v >> 8)
-    p[2] = <uint8_t>(v >> 16)
-    p[3] = <uint8_t>(v >> 24)
-
-
 cdef inline uint64_t u8to64_le(uint8_t* p) nogil:
     return (<uint64_t>p[0] |
             <uint64_t>p[1] << 8 |

pandas/_libs/hashing.pyx

@@ -1046,7 +1046,7 @@ cdef class StringHashTable(HashTable):
             not None, then _additionally_ any value "val" satisfying
             val == na_value is considered missing.
         mask : ndarray[bool], optional
-            Not yet implementd for StringHashTable.
+            Not yet implemented for StringHashTable.
 
         Returns
         -------

pandas/_libs/hashtable_class_helper.pxi.in

@@ -62,4 +62,4 @@ class BaseMultiIndexCodesEngine:
         values: np.ndarray,  # np.ndarray[object] of tuples
         method: str,
         limit: int | None,
-    ) -> np.ndarray: ...  # np.ndarray[np.int64]
+    ) -> npt.NDArray[np.intp]: ...

pandas/_libs/index.pyi

@@ -13,7 +13,10 @@
     algos as libalgos,
     lib,
 )
-from pandas._typing import ArrayLike
+from pandas._typing import (
+    ArrayLike,
+    npt,
+)
 
 from pandas.core.dtypes.cast import maybe_promote
 from pandas.core.dtypes.common import (
@@ -110,7 +113,7 @@ def take_nd(
 
 def _take_nd_ndarray(
     arr: np.ndarray,
-    indexer,
+    indexer: npt.NDArray[np.intp] | None,
     axis: int,
     fill_value,
     allow_fill: bool,
@@ -122,7 +125,7 @@ def _take_nd_ndarray(
     else:
         indexer = ensure_platform_int(indexer)
 
-    indexer, dtype, fill_value, mask_info = _take_preprocess_indexer_and_fill_value(
+    dtype, fill_value, mask_info = _take_preprocess_indexer_and_fill_value(
         arr, indexer, fill_value, allow_fill
     )
 
@@ -160,30 +163,32 @@ def _take_nd_ndarray(
 
 def take_1d(
     arr: ArrayLike,
-    indexer: np.ndarray,
+    indexer: npt.NDArray[np.intp],
     fill_value=None,
     allow_fill: bool = True,
 ) -> ArrayLike:
     """
     Specialized version for 1D arrays. Differences compared to `take_nd`:
 
     - Assumes input array has already been converted to numpy array / EA
-    - Assumes indexer is already guaranteed to be int64 dtype ndarray
+    - Assumes indexer is already guaranteed to be intp dtype ndarray
     - Only works for 1D arrays
 
     To ensure the lowest possible overhead.
 
     Note: similarly to `take_nd`, this function assumes that the indexer is
     a valid(ated) indexer with no out of bound indices.
     """
+    indexer = ensure_platform_int(indexer)
+
     if not isinstance(arr, np.ndarray):
         # ExtensionArray -> dispatch to their method
         return arr.take(indexer, fill_value=fill_value, allow_fill=allow_fill)
 
     if not allow_fill:
         return arr.take(indexer)
 
-    indexer, dtype, fill_value, mask_info = _take_preprocess_indexer_and_fill_value(
+    dtype, fill_value, mask_info = _take_preprocess_indexer_and_fill_value(
         arr, indexer, fill_value, True
     )
 
@@ -200,7 +205,9 @@ def take_1d(
 
 
 def take_2d_multi(
-    arr: np.ndarray, indexer: tuple[np.ndarray, np.ndarray], fill_value=np.nan
+    arr: np.ndarray,
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
+    fill_value=np.nan,
 ) -> np.ndarray:
     """
     Specialized Cython take which sets NaN values in one pass.
@@ -249,6 +256,7 @@ def take_2d_multi(
     if func is not None:
         func(arr, indexer, out=out, fill_value=fill_value)
     else:
+        # test_reindex_multi
         _take_2d_multi_object(
             arr, indexer, out, fill_value=fill_value, mask_info=mask_info
         )
@@ -457,7 +465,7 @@ def wrapper(
 
 def _take_nd_object(
     arr: np.ndarray,
-    indexer: np.ndarray,  # np.ndarray[np.intp]
+    indexer: npt.NDArray[np.intp],
     out: np.ndarray,
     axis: int,
     fill_value,
@@ -480,7 +488,7 @@ def _take_nd_object(
 
 def _take_2d_multi_object(
     arr: np.ndarray,
-    indexer: tuple[np.ndarray, np.ndarray],
+    indexer: tuple[npt.NDArray[np.intp], npt.NDArray[np.intp]],
     out: np.ndarray,
     fill_value,
     mask_info,
@@ -509,7 +517,7 @@ def _take_2d_multi_object(
 
 def _take_preprocess_indexer_and_fill_value(
     arr: np.ndarray,
-    indexer: np.ndarray,
+    indexer: npt.NDArray[np.intp],
     fill_value,
     allow_fill: bool,
 ):
@@ -533,5 +541,4 @@ def _take_preprocess_indexer_and_fill_value(
                 # to crash when trying to cast it to dtype)
                 dtype, fill_value = arr.dtype, arr.dtype.type()
 
-    indexer = ensure_platform_int(indexer)
-    return indexer, dtype, fill_value, mask_info
+    return dtype, fill_value, mask_info

pandas/core/array_algos/take.py

@@ -4672,17 +4672,23 @@ def _reindex_columns(
             allow_dups=False,
         )
 
-    def _reindex_multi(self, axes, copy: bool, fill_value) -> DataFrame:
+    def _reindex_multi(
+        self, axes: dict[str, Index], copy: bool, fill_value
+    ) -> DataFrame:
         """
         We are guaranteed non-Nones in the axes.
         """
+
         new_index, row_indexer = self.index.reindex(axes["index"])
         new_columns, col_indexer = self.columns.reindex(axes["columns"])
 
         if row_indexer is not None and col_indexer is not None:
+            # Fastpath. By doing two 'take's at once we avoid making an
+            #  unnecessary copy.
+            # We only get here with `not self._is_mixed_type`, which (almost)
+            #  ensures that self.values is cheap. It may be worth making this
+            #  condition more specific.
             indexer = row_indexer, col_indexer
-            # error: Argument 2 to "take_2d_multi" has incompatible type "Tuple[Any,
-            # Any]"; expected "ndarray"
             new_values = take_2d_multi(self.values, indexer, fill_value=fill_value)
             return self._constructor(new_values, index=new_index, columns=new_columns)
         else:

pandas/core/frame.py

@@ -31,7 +31,7 @@
 _T = TypeVar("_T", bound="NDArrayBackedExtensionIndex")
 
 
-def inherit_from_data(name: str, delegate, cache: bool = False, wrap: bool = False):
+def _inherit_from_data(name: str, delegate, cache: bool = False, wrap: bool = False):
     """
     Make an alias for a method of the underlying ExtensionArray.
 
@@ -120,7 +120,7 @@ def inherit_names(names: list[str], delegate, cache: bool = False, wrap: bool =
 
     def wrapper(cls):
         for name in names:
-            meth = inherit_from_data(name, delegate, cache=cache, wrap=wrap)
+            meth = _inherit_from_data(name, delegate, cache=cache, wrap=wrap)
             setattr(cls, name, meth)
 
         return cls

pandas/core/indexes/extension.py

@@ -774,7 +774,7 @@ def test_apply_series_on_date_time_index_aware_series(dti, exp, aware):
     tm.assert_frame_equal(result, exp)
 
 
-def test_apply_scaler_on_date_time_index_aware_series():
+def test_apply_scalar_on_date_time_index_aware_series():
     # GH 25959
     # Calling apply on a localized time series should not cause an error
     series = tm.makeTimeSeries(nper=30).tz_localize("UTC")