Backport PR #37473: CI: 32 bit maybe_indices_to_slice (#37503)

Co-authored-by: jbrockmendel <jbrockmendel@gmail.com>

Author: meeseeksmachine

pandas/_libs/lib.pyx

@@ -36,6 +36,7 @@ from numpy cimport (
     float32_t,
     float64_t,
     int64_t,
+    intp_t,
     ndarray,
     uint8_t,
     uint64_t,
@@ -490,7 +491,7 @@ def has_infs_f8(const float64_t[:] arr) -> bool:
     return False
 
 
-def maybe_indices_to_slice(ndarray[int64_t] indices, int max_len):
+def maybe_indices_to_slice(ndarray[intp_t] indices, int max_len):
     cdef:
         Py_ssize_t i, n = len(indices)
         int k, vstart, vlast, v

pandas/core/frame.py

@@ -2881,7 +2881,9 @@ def __getitem__(self, key):
         indexer = convert_to_index_sliceable(self, key)
         if indexer is not None:
             if isinstance(indexer, np.ndarray):
-                indexer = lib.maybe_indices_to_slice(indexer, len(self))
+                indexer = lib.maybe_indices_to_slice(
+                    indexer.astype(np.intp, copy=False), len(self)
+                )
             # either we have a slice or we have a string that can be converted
             #  to a slice for partial-string date indexing
             return self._slice(indexer, axis=0)

pandas/core/indexes/base.py

@@ -5094,7 +5094,9 @@ def get_slice_bound(self, label, side: str_t, kind) -> int:
             if is_bool_dtype(slc):
                 slc = lib.maybe_booleans_to_slice(slc.view("u1"))
             else:
-                slc = lib.maybe_indices_to_slice(slc.astype("i8"), len(self))
+                slc = lib.maybe_indices_to_slice(
+                    slc.astype(np.intp, copy=False), len(self)
+                )
             if isinstance(slc, np.ndarray):
                 raise KeyError(
                     f"Cannot get {side} slice bound for non-unique "

pandas/core/indexes/datetimelike.py

@@ -15,7 +15,6 @@
 from pandas.util._decorators import Appender, cache_readonly, doc
 
 from pandas.core.dtypes.common import (
-    ensure_int64,
     is_bool_dtype,
     is_dtype_equal,
     is_integer,
@@ -181,7 +180,7 @@ def sort_values(self, return_indexer=False, ascending=True, key=None):
     @Appender(_index_shared_docs["take"] % _index_doc_kwargs)
     def take(self, indices, axis=0, allow_fill=True, fill_value=None, **kwargs):
         nv.validate_take(tuple(), kwargs)
-        indices = ensure_int64(indices)
+        indices = np.asarray(indices, dtype=np.intp)
 
         maybe_slice = lib.maybe_indices_to_slice(indices, len(self))
         if isinstance(maybe_slice, slice):
@@ -581,7 +580,9 @@ def delete(self, loc):
                 freq = self.freq
         else:
             if is_list_like(loc):
-                loc = lib.maybe_indices_to_slice(ensure_int64(np.array(loc)), len(self))
+                loc = lib.maybe_indices_to_slice(
+                    np.asarray(loc, dtype=np.intp), len(self)
+                )
             if isinstance(loc, slice) and loc.step in (1, None):
                 if loc.start in (0, None) or loc.stop in (len(self), None):
                     freq = self.freq

pandas/core/indexes/multi.py

@@ -2688,7 +2688,7 @@ def get_loc(self, key, method=None):
 
         def _maybe_to_slice(loc):
             """convert integer indexer to boolean mask or slice if possible"""
-            if not isinstance(loc, np.ndarray) or loc.dtype != "int64":
+            if not isinstance(loc, np.ndarray) or loc.dtype != np.intp:
                 return loc
 
             loc = lib.maybe_indices_to_slice(loc, len(self))
@@ -2735,7 +2735,7 @@ def _maybe_to_slice(loc):
             stacklevel=10,
         )
 
-        loc = np.arange(start, stop, dtype="int64")
+        loc = np.arange(start, stop, dtype=np.intp)
 
         for i, k in enumerate(follow_key, len(lead_key)):
             mask = self.codes[i][loc] == self._get_loc_single_level_index(

pandas/core/internals/managers.py

@@ -239,8 +239,8 @@ def _rebuild_blknos_and_blklocs(self) -> None:
         """
         Update mgr._blknos / mgr._blklocs.
         """
-        new_blknos = np.empty(self.shape[0], dtype=np.int64)
-        new_blklocs = np.empty(self.shape[0], dtype=np.int64)
+        new_blknos = np.empty(self.shape[0], dtype=np.intp)
+        new_blklocs = np.empty(self.shape[0], dtype=np.intp)
         new_blknos.fill(-1)
         new_blklocs.fill(-1)
 

pandas/tests/test_lib.py

@@ -51,15 +51,15 @@ def test_maybe_indices_to_slice_left_edge(self):
         target = np.arange(100)
 
         # slice
-        indices = np.array([], dtype=np.int64)
+        indices = np.array([], dtype=np.intp)
         maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
 
         assert isinstance(maybe_slice, slice)
         tm.assert_numpy_array_equal(target[indices], target[maybe_slice])
 
         for end in [1, 2, 5, 20, 99]:
             for step in [1, 2, 4]:
-                indices = np.arange(0, end, step, dtype=np.int64)
+                indices = np.arange(0, end, step, dtype=np.intp)
                 maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
 
                 assert isinstance(maybe_slice, slice)
@@ -74,7 +74,7 @@ def test_maybe_indices_to_slice_left_edge(self):
 
         # not slice
         for case in [[2, 1, 2, 0], [2, 2, 1, 0], [0, 1, 2, 1], [-2, 0, 2], [2, 0, -2]]:
-            indices = np.array(case, dtype=np.int64)
+            indices = np.array(case, dtype=np.intp)
             maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
 
             assert not isinstance(maybe_slice, slice)
@@ -87,7 +87,7 @@ def test_maybe_indices_to_slice_right_edge(self):
         # slice
         for start in [0, 2, 5, 20, 97, 98]:
             for step in [1, 2, 4]:
-                indices = np.arange(start, 99, step, dtype=np.int64)
+                indices = np.arange(start, 99, step, dtype=np.intp)
                 maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
 
                 assert isinstance(maybe_slice, slice)
@@ -101,7 +101,7 @@ def test_maybe_indices_to_slice_right_edge(self):
                 tm.assert_numpy_array_equal(target[indices], target[maybe_slice])
 
         # not slice
-        indices = np.array([97, 98, 99, 100], dtype=np.int64)
+        indices = np.array([97, 98, 99, 100], dtype=np.intp)
         maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
 
         assert not isinstance(maybe_slice, slice)
@@ -114,7 +114,7 @@ def test_maybe_indices_to_slice_right_edge(self):
         with pytest.raises(IndexError, match=msg):
             target[maybe_slice]
 
-        indices = np.array([100, 99, 98, 97], dtype=np.int64)
+        indices = np.array([100, 99, 98, 97], dtype=np.intp)
         maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
 
         assert not isinstance(maybe_slice, slice)
@@ -126,7 +126,7 @@ def test_maybe_indices_to_slice_right_edge(self):
             target[maybe_slice]
 
         for case in [[99, 97, 99, 96], [99, 99, 98, 97], [98, 98, 97, 96]]:
-            indices = np.array(case, dtype=np.int64)
+            indices = np.array(case, dtype=np.intp)
             maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
 
             assert not isinstance(maybe_slice, slice)
@@ -138,7 +138,7 @@ def test_maybe_indices_to_slice_both_edges(self):
 
         # slice
         for step in [1, 2, 4, 5, 8, 9]:
-            indices = np.arange(0, 9, step, dtype=np.int64)
+            indices = np.arange(0, 9, step, dtype=np.intp)
             maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
             assert isinstance(maybe_slice, slice)
             tm.assert_numpy_array_equal(target[indices], target[maybe_slice])
@@ -151,7 +151,7 @@ def test_maybe_indices_to_slice_both_edges(self):
 
         # not slice
         for case in [[4, 2, 0, -2], [2, 2, 1, 0], [0, 1, 2, 1]]:
-            indices = np.array(case, dtype=np.int64)
+            indices = np.array(case, dtype=np.intp)
             maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
             assert not isinstance(maybe_slice, slice)
             tm.assert_numpy_array_equal(maybe_slice, indices)
@@ -163,7 +163,7 @@ def test_maybe_indices_to_slice_middle(self):
         # slice
         for start, end in [(2, 10), (5, 25), (65, 97)]:
             for step in [1, 2, 4, 20]:
-                indices = np.arange(start, end, step, dtype=np.int64)
+                indices = np.arange(start, end, step, dtype=np.intp)
                 maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
 
                 assert isinstance(maybe_slice, slice)
@@ -178,7 +178,7 @@ def test_maybe_indices_to_slice_middle(self):
 
         # not slice
         for case in [[14, 12, 10, 12], [12, 12, 11, 10], [10, 11, 12, 11]]:
-            indices = np.array(case, dtype=np.int64)
+            indices = np.array(case, dtype=np.intp)
             maybe_slice = lib.maybe_indices_to_slice(indices, len(target))
 
             assert not isinstance(maybe_slice, slice)