BUG: hashing read only object categories raises (#58481)

pandas-dev · Apr 30, 2024 · 59f6a33 · 59f6a33
1 parent 086b047
commit 59f6a33
Show file tree

Hide file tree

Showing 2 changed files with 10 additions and 1 deletion.
diff --git a/pandas/_libs/hashing.pyx b/pandas/_libs/hashing.pyx
@@ -11,6 +11,7 @@ import numpy as np
 
 from numpy cimport (
     import_array,
+    ndarray,
     uint8_t,
     uint64_t,
 )
@@ -22,7 +23,7 @@ from pandas._libs.util cimport is_nan
 
 @cython.boundscheck(False)
 def hash_object_array(
-    object[:] arr, str key, str encoding="utf8"
+    ndarray[object, ndim=1] arr, str key, str encoding="utf8"
 ) -> np.ndarray[np.uint64]:
     """
     Parameters

diff --git a/pandas/tests/arrays/categorical/test_algos.py b/pandas/tests/arrays/categorical/test_algos.py
@@ -86,3 +86,11 @@ def test_diff():
     df = ser.to_frame(name="A")
     with pytest.raises(TypeError, match=msg):
         df.diff()
+
+
+def test_hash_read_only_categorical():
+    # GH#58481
+    idx = pd.Index(pd.Index(["a", "b", "c"], dtype="object").values)
+    cat = pd.CategoricalDtype(idx)
+    arr = pd.Series(["a", "b"], dtype=cat).values
+    assert hash(arr.dtype) == hash(arr.dtype)