From b39022ae4b96a4a83300551c0b61008dd75fa9eb Mon Sep 17 00:00:00 2001
From: Matthew Roeschke <10647082+mroeschke@users.noreply.github.com>
Date: Thu, 21 Sep 2023 10:43:25 -0700
Subject: [PATCH 1/2] Fix DataFrame from Series with different
 CategoricalIndexes

---
 python/cudf/cudf/core/indexed_frame.py   |  7 +++++++
 python/cudf/cudf/tests/test_dataframe.py | 13 +++++++++++++
 2 files changed, 20 insertions(+)

diff --git a/python/cudf/cudf/core/indexed_frame.py b/python/cudf/cudf/core/indexed_frame.py
index 62e091b29b5..aacf1fa8dae 100644
--- a/python/cudf/cudf/core/indexed_frame.py
+++ b/python/cudf/cudf/core/indexed_frame.py
@@ -5438,6 +5438,13 @@ def _is_same_dtype(lhs_dtype, rhs_dtype):
     # for matching column dtype.
     if lhs_dtype == rhs_dtype:
         return True
+    elif (
+        is_categorical_dtype(lhs_dtype)
+        and is_categorical_dtype(rhs_dtype)
+        and lhs_dtype.categories.dtype == rhs_dtype.categories.dtype
+    ):
+        # OK if categories are not all the same
+        return True
     elif (
         is_categorical_dtype(lhs_dtype)
         and not is_categorical_dtype(rhs_dtype)
diff --git a/python/cudf/cudf/tests/test_dataframe.py b/python/cudf/cudf/tests/test_dataframe.py
index 6180162ecdd..61c3c4deffd 100644
--- a/python/cudf/cudf/tests/test_dataframe.py
+++ b/python/cudf/cudf/tests/test_dataframe.py
@@ -10389,6 +10389,19 @@ def test_dataframe_init_from_nested_dict():
     assert_eq(pdf, gdf)
 
 
+def test_init_from_2_categoricalindex_series_diff_categories():
+    s1 = cudf.Series(
+        [39, 6, 4], index=cudf.CategoricalIndex(["female", "male", "unknown"])
+    )
+    s2 = cudf.Series(
+        [2, 152, 2, 242, 150],
+        index=cudf.CategoricalIndex(["f", "female", "m", "male", "unknown"]),
+    )
+    result = cudf.DataFrame([s1, s2])
+    expected = pd.DataFrame([s1.to_pandas(), s2.to_pandas()])
+    assert_eq(result, expected)
+
+
 def test_data_frame_values_no_cols_but_index():
     result = cudf.DataFrame(index=range(5)).values
     expected = pd.DataFrame(index=range(5)).values

From a5bb2594a45127dccb4faf96c50bcdb2d58e0fcb Mon Sep 17 00:00:00 2001
From: Matthew Roeschke <10647082+mroeschke@users.noreply.github.com>
Date: Fri, 22 Sep 2023 10:29:37 -0700
Subject: [PATCH 2/2] check_dtype=False

---
 python/cudf/cudf/tests/test_dataframe.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/python/cudf/cudf/tests/test_dataframe.py b/python/cudf/cudf/tests/test_dataframe.py
index 6dce2990851..67b63028fab 100644
--- a/python/cudf/cudf/tests/test_dataframe.py
+++ b/python/cudf/cudf/tests/test_dataframe.py
@@ -10418,7 +10418,7 @@ def test_init_from_2_categoricalindex_series_diff_categories():
     )
     result = cudf.DataFrame([s1, s2])
     expected = pd.DataFrame([s1.to_pandas(), s2.to_pandas()])
-    assert_eq(result, expected)
+    assert_eq(result, expected, check_dtype=False)
 
 
 def test_data_frame_values_no_cols_but_index():