BUG: fix to_datetime failing to raise on mixed tznaive/tzaware datetimes (#24663)

Author: jbrockmendel

doc/source/whatsnew/v0.24.0.rst

@@ -1551,6 +1551,7 @@ Datetimelike
 - Bug in :class:`PeriodIndex` where comparisons against an array-like object with length 1 failed to raise ``ValueError`` (:issue:`23078`)
 - Bug in :meth:`DatetimeIndex.astype`, :meth:`PeriodIndex.astype` and :meth:`TimedeltaIndex.astype` ignoring the sign of the ``dtype`` for unsigned integer dtypes (:issue:`24405`).
 - Fixed bug in :meth:`Series.max` with ``datetime64[ns]``-dtype failing to return ``NaT`` when nulls are present and ``skipna=False`` is passed (:issue:`24265`)
+- Bug in :func:`to_datetime` where arrays of ``datetime`` objects containing both timezone-aware and timezone-naive ``datetimes`` would fail to raise ``ValueError`` (:issue:`24569`)
 
 Timedelta
 ^^^^^^^^^

pandas/_libs/tslibs/conversion.pyx

@@ -167,6 +167,7 @@ def datetime_to_datetime64(values: object[:]):
         int64_t[:] iresult
         npy_datetimestruct dts
         _TSObject _ts
+        bint found_naive = False
 
     result = np.empty(n, dtype='M8[ns]')
     iresult = result.view('i8')
@@ -176,6 +177,9 @@ def datetime_to_datetime64(values: object[:]):
             iresult[i] = NPY_NAT
         elif PyDateTime_Check(val):
             if val.tzinfo is not None:
+                if found_naive:
+                    raise ValueError('Cannot mix tz-aware with '
+                                     'tz-naive values')
                 if inferred_tz is not None:
                     if not tz_compare(val.tzinfo, inferred_tz):
                         raise ValueError('Array must be all same time zone')
@@ -186,6 +190,7 @@ def datetime_to_datetime64(values: object[:]):
                 iresult[i] = _ts.value
                 check_dts_bounds(&_ts.dts)
             else:
+                found_naive = True
                 if inferred_tz is not None:
                     raise ValueError('Cannot mix tz-aware with '
                                      'tz-naive values')

pandas/tests/arrays/test_array.py

@@ -179,11 +179,7 @@ def test_array_inference(data, expected):
     [pd.Timestamp("2000", tz="CET"), pd.Timestamp("2000", tz="UTC")],
     # Mix of tz-aware and tz-naive
     [pd.Timestamp("2000", tz="CET"), pd.Timestamp("2000")],
-    # GH-24569
-    pytest.param(
-        np.array([pd.Timestamp('2000'), pd.Timestamp('2000', tz='CET')]),
-        marks=pytest.mark.xfail(reason="bug in DTA._from_sequence")
-    ),
+    np.array([pd.Timestamp('2000'), pd.Timestamp('2000', tz='CET')]),
 ])
 def test_array_inference_fails(data):
     result = pd.array(data)

pandas/tests/arrays/test_datetimes.py

@@ -16,6 +16,24 @@
 
 
 class TestDatetimeArrayConstructor(object):
+    @pytest.mark.parametrize('meth', [DatetimeArray._from_sequence,
+                                      sequence_to_dt64ns,
+                                      pd.to_datetime,
+                                      pd.DatetimeIndex])
+    def test_mixing_naive_tzaware_raises(self, meth):
+        # GH#24569
+        arr = np.array([pd.Timestamp('2000'), pd.Timestamp('2000', tz='CET')])
+
+        msg = ('Cannot mix tz-aware with tz-naive values|'
+               'Tz-aware datetime.datetime cannot be converted '
+               'to datetime64 unless utc=True')
+
+        for obj in [arr, arr[::-1]]:
+            # check that we raise regardless of whether naive is found
+            #  before aware or vice-versa
+            with pytest.raises(ValueError, match=msg):
+                meth(obj)
+
     def test_from_pandas_array(self):
         arr = pd.array(np.arange(5, dtype=np.int64)) * 3600 * 10**9
 

pandas/tests/indexes/datetimes/test_construction.py

@@ -306,24 +306,15 @@ def test_construction_dti_with_mixed_timezones(self):
         tm.assert_index_equal(result, exp, exact=True)
         assert isinstance(result, DatetimeIndex)
 
-        # different tz coerces tz-naive to tz-awareIndex(dtype=object)
-        result = DatetimeIndex([Timestamp('2011-01-01 10:00'),
-                                Timestamp('2011-01-02 10:00',
-                                          tz='US/Eastern')], name='idx')
-        exp = DatetimeIndex([Timestamp('2011-01-01 05:00'),
-                             Timestamp('2011-01-02 10:00')],
-                            tz='US/Eastern', name='idx')
-        tm.assert_index_equal(result, exp, exact=True)
-        assert isinstance(result, DatetimeIndex)
-
         # tz mismatch affecting to tz-aware raises TypeError/ValueError
 
         with pytest.raises(ValueError):
             DatetimeIndex([Timestamp('2011-01-01 10:00', tz='Asia/Tokyo'),
                            Timestamp('2011-01-02 10:00', tz='US/Eastern')],
                           name='idx')
 
-        with pytest.raises(TypeError, match='data is already tz-aware'):
+        msg = 'cannot be converted to datetime64'
+        with pytest.raises(ValueError, match=msg):
             DatetimeIndex([Timestamp('2011-01-01 10:00'),
                            Timestamp('2011-01-02 10:00', tz='US/Eastern')],
                           tz='Asia/Tokyo', name='idx')
@@ -333,7 +324,7 @@ def test_construction_dti_with_mixed_timezones(self):
                            Timestamp('2011-01-02 10:00', tz='US/Eastern')],
                           tz='US/Eastern', name='idx')
 
-        with pytest.raises(TypeError, match='data is already tz-aware'):
+        with pytest.raises(ValueError, match=msg):
             # passing tz should results in DatetimeIndex, then mismatch raises
             # TypeError
             Index([pd.NaT, Timestamp('2011-01-01 10:00'),