Raise error in reindex when index is not unique

python/cudf/cudf/core/indexed_frame.py

@@ -2606,6 +2606,10 @@ def _reindex(
 
         df = self
         if index is not None:
+            if not df._index.is_unique:
+                raise ValueError(
+                    "cannot reindex on an axis with duplicate labels"
+                )
             index = cudf.core.index.as_index(
                 index, name=getattr(index, "name", self._index.name)
             )

python/cudf/cudf/tests/test_dataframe.py

@@ -10723,3 +10723,15 @@ def test_dataframe_series_dot():
     expected = gser @ [12, 13]
 
     assert_eq(expected, actual)
+
+
+def test_dataframe_duplicate_index_reindex():
+    gdf = cudf.DataFrame({"a": [0, 1, 2, 3]}, index=[0, 0, 1, 1])
+    pdf = gdf.to_pandas()
+
+    assert_exceptions_equal(
+        gdf.reindex,
+        pdf.reindex,
+        lfunc_args_and_kwargs=([10, 11, 12, 13], {}),
+        rfunc_args_and_kwargs=([10, 11, 12, 13], {}),
+    )

python/cudf/cudf/tests/test_series.py

@@ -2638,3 +2638,15 @@ def test_series_setitem_mixed_bool_dtype():
     s = cudf.Series([True, False, True])
     with pytest.raises(TypeError):
         s[0] = 10
+
+
+def test_series_duplicate_index_reindex():
+    gs = cudf.Series([0, 1, 2, 3], index=[0, 0, 1, 1])
+    ps = gs.to_pandas()
+
+    assert_exceptions_equal(
+        gs.reindex,
+        ps.reindex,
+        lfunc_args_and_kwargs=([10, 11, 12, 13], {}),
+        rfunc_args_and_kwargs=([10, 11, 12, 13], {}),
+    )

python/dask_cudf/dask_cudf/backends.py

@@ -437,28 +437,23 @@ def union_categoricals_cudf(
     )
 
 
-@_dask_cudf_nvtx_annotate
-def safe_hash(frame):
-    return cudf.Series(frame.hash_values(), index=frame.index)
-
-
 @hash_object_dispatch.register((cudf.DataFrame, cudf.Series))
 @_dask_cudf_nvtx_annotate
 def hash_object_cudf(frame, index=True):
     if index:
-        return safe_hash(frame.reset_index())
-    return safe_hash(frame)
+        frame = frame.reset_index()
+    return frame.hash_values()
 
 
 @hash_object_dispatch.register(cudf.BaseIndex)
 @_dask_cudf_nvtx_annotate
 def hash_object_cudf_index(ind, index=None):
 
     if isinstance(ind, cudf.MultiIndex):
-        return safe_hash(ind.to_frame(index=False))
+        return ind.to_frame(index=False).hash_values()
 
     col = cudf.core.column.as_column(ind)
-    return safe_hash(cudf.Series(col))
+    return cudf.Series(col).hash_values()
 
 
 @group_split_dispatch.register((cudf.Series, cudf.DataFrame))