BUG: caught typeError in series.at (#25506)

doc/source/whatsnew/v0.25.0.rst

@@ -150,7 +150,7 @@ Timezones
 
 - Bug in :func:`to_datetime` with ``utc=True`` and datetime strings that would apply previously parsed UTC offsets to subsequent arguments (:issue:`24992`)
 - Bug in :func:`Timestamp.tz_localize` and :func:`Timestamp.tz_convert` does not propagate ``freq`` (:issue:`25241`)
--
+- Bug in :func:`Series.at` where setting :class:`Timestamp` with timezone raises ``TypeError`` (:issue:`25506`)
 
 Numeric
 ^^^^^^^

pandas/core/series.py

@@ -1229,7 +1229,7 @@ def _set_value(self, label, value, takeable=False):
                 self._values[label] = value
             else:
                 self.index._engine.set_value(self._values, label, value)
-        except KeyError:
+        except (KeyError, TypeError):
 
             # set using a non-recursive method
             self.loc[label] = value

pandas/tests/indexing/test_scalar.py

@@ -185,6 +185,14 @@ def test_at_with_tz(self):
         result = df.at[0, 'date']
         assert result == expected
 
+    def test_series_set_tz_timestamp(self, tz_naive_fixture):
+        # GH 25506
+        ts = Timestamp('2017-08-05 00:00:00+0100', tz=tz_naive_fixture)
+        result = Series(ts)
+        result.at[1] = ts
+        expected = Series([ts, ts])
+        tm.assert_series_equal(result, expected)
+
     def test_mixed_index_at_iat_loc_iloc_series(self):
         # GH 19860
         s = Series([1, 2, 3, 4, 5], index=['a', 'b', 'c', 1, 2])