BUG: Fix dataframe.update not updating dtype #55509 v2

doc/source/whatsnew/v3.0.0.rst

@@ -264,6 +264,7 @@ Bug fixes
 ~~~~~~~~~
 - Fixed bug in :meth:`DataFrame.join` inconsistently setting result index name (:issue:`55815`)
 - Fixed bug in :meth:`DataFrame.to_string` that raised ``StopIteration`` with nested DataFrames. (:issue:`16098`)
+- Fixed bug in :meth:`DataFrame.update` bool dtype being converted to object (:issue:`55509`)
 - Fixed bug in :meth:`Series.diff` allowing non-integer values for the ``periods`` argument. (:issue:`56607`)
 
 Categorical

pandas/core/frame.py

@@ -8692,6 +8692,10 @@ def update(
         dict.update : Similar method for dictionaries.
         DataFrame.merge : For column(s)-on-column(s) operations.
 
+        Notes
+        --------
+        1. Duplicate indices on `other` are not supported and raises `ValueError`.
+
         Examples
         --------
         >>> df = pd.DataFrame({"A": [1, 2, 3], "B": [400, 500, 600]})
@@ -8764,11 +8768,22 @@ def update(
         if not isinstance(other, DataFrame):
             other = DataFrame(other)
 
-        other = other.reindex(self.index)
+        if other.index.has_duplicates:
+            raise ValueError("Update not allowed with duplicate indexes on other.")
+
+        rows = other.index.intersection(self.index)
+        if rows.empty:
+            raise ValueError(
+                "Can't update dataframe when other has no index in common with "
+                "this dataframe."
+            )
+
+        other = other.reindex(rows)
+        this_data = self.loc[rows]
 
         for col in self.columns.intersection(other.columns):
-            this = self[col]._values
-            that = other[col]._values
+            this = this_data[col]
+            that = other[col]
 
             if filter_func is not None:
                 mask = ~filter_func(this) | isna(that)
@@ -8788,7 +8803,7 @@ def update(
             if mask.all():
                 continue
 
-            self.loc[:, col] = self[col].where(mask, that)
+            self.loc[rows, col] = this.where(mask, that)
 
     # ----------------------------------------------------------------------
     # Data reshaping

pandas/tests/frame/methods/test_update.py

@@ -184,3 +184,47 @@ def test_update_dt_column_with_NaT_create_column(self):
             {"A": [1.0, 3.0], "B": [pd.NaT, pd.to_datetime("2016-01-01")]}
         )
         tm.assert_frame_equal(df, expected)
+
+    @pytest.mark.parametrize(
+        "value_df, value_other, dtype",
+        [
+            (True, False, bool),
+            (1, 2, int),
+            (np.uint64(1), np.uint(2), np.dtype("uint64")),
+            (1.0, 2.0, float),
+            (1.0 + 1j, 2.0 + 2j, complex),
+            ("a", "b", pd.StringDtype()),
+            (
+                pd.to_timedelta("1 ms"),
+                pd.to_timedelta("2 ms"),
+                np.dtype("timedelta64[ns]"),
+            ),
+            (
+                np.datetime64("2000-01-01T00:00:00"),
+                np.datetime64("2000-01-02T00:00:00"),
+                np.dtype("datetime64[ns]"),
+            ),
+        ],
+    )
+    def test_update_preserve_dtype(self, value_df, value_other, dtype):
+        # GH#55509
+        df = DataFrame({"a": [value_df] * 2}, index=[1, 2])
+        other = DataFrame({"a": [value_other]}, index=[1])
+        expected = DataFrame({"a": [value_other, value_df]}, index=[1, 2])
+        df.update(other)
+        tm.assert_frame_equal(df, expected)
+
+    def test_update_raises_on_duplicate_argument_index(self):
+        # GH#55509
+        df = DataFrame({"a": [1, 1]}, index=[1, 2])
+        other = DataFrame({"a": [2, 3]}, index=[1, 1])
+        with pytest.raises(ValueError, match="duplicate index"):
+            df.update(other)
+
+    def test_update_on_duplicate_frame_unique_argument_index(self):
+        # GH#55509
+        df = DataFrame({"a": [1, 1, 1]}, index=[1, 1, 2])
+        other = DataFrame({"a": [2, 3]}, index=[1, 2])
+        expected = DataFrame({"a": [2, 2, 3]}, index=[1, 1, 2])
+        df.update(other)
+        tm.assert_frame_equal(df, expected)