ENH: Format datetime.datetime and pd.Timestamp objects in pd.to_datetime

doc/source/whatsnew/v2.0.0.rst

@@ -49,6 +49,7 @@ Other enhancements
 - Fix ``test`` optional_extra by adding missing test package ``pytest-asyncio`` (:issue:`48361`)
 - :func:`DataFrame.astype` exception message thrown improved to include column name when type conversion is not possible. (:issue:`47571`)
 - :meth:`DataFrame.to_json` now supports a ``mode`` keyword with supported inputs 'w' and 'a'. Defaulting to 'w', 'a' can be used when lines=True and orient='records' to append record oriented json lines to an existing json file. (:issue:`35849`)
+- :func:`to_datetime` now correctly parses ``datetime.datetime`` objects in the input when using the ``format`` argument instead of raising a ``ValueError``. (:issue:`49298`)
 
 .. ---------------------------------------------------------------------------
 .. _whatsnew_200.notable_bug_fixes:

pandas/_libs/tslibs/strptime.pyx

@@ -2,6 +2,7 @@
 """
 from cpython.datetime cimport (
     date,
+    datetime,
     tzinfo,
 )
 
@@ -129,12 +130,22 @@ def array_strptime(ndarray[object] values, str fmt, bint exact=True, errors='rai
             if val in nat_strings:
                 iresult[i] = NPY_NAT
                 continue
+        elif checknull_with_nat_and_na(val):
+            iresult[i] = NPY_NAT
+            continue
+        elif isinstance(val, datetime):
+            iresult[i] = _parse_python_datetime_object(val, &dts)
+            try:
+                check_dts_bounds(&dts)
+            except ValueError:
+                if is_coerce:
+                    iresult[i] = NPY_NAT
+                    continue
+                raise
+            result_timezone[i] = val.tzname()
+            continue
         else:
-            if checknull_with_nat_and_na(val):
-                iresult[i] = NPY_NAT
-                continue
-            else:
-                val = str(val)
+            val = str(val)
 
         # exact matching
         if exact:
@@ -532,3 +543,29 @@ cdef tzinfo parse_timezone_directive(str z):
                      (microseconds // 60_000_000))
     total_minutes = -total_minutes if z.startswith("-") else total_minutes
     return pytz.FixedOffset(total_minutes)
+
+cdef int64_t _parse_python_datetime_object(datetime dt, npy_datetimestruct *dts):
+    """
+    Parse a native datetime.datetime object and return a numpy datetime object
+
+    Parameters
+    ----------
+    dt : datetime.datetime instance
+    dts: numpy datetime struct
+
+    Returns
+    -------
+    int64_t
+        the numpy datetime object
+    """
+    dts.year = dt.year
+    dts.month = dt.month
+    dts.day = dt.day
+    dts.hour = dt.hour
+    dts.min = dt.minute
+    dts.sec = dt.second
+    dts.us = dt.microsecond
+    dts.ps = 0  # Not enough precision in datetime objects (https://github.com/python/cpython/issues/59648)
+
+    npy_datetime = npy_datetimestruct_to_datetime(NPY_FR_ns, dts)
+    return npy_datetime

pandas/tests/tools/test_to_datetime.py

@@ -609,6 +609,56 @@ def test_to_datetime_dtarr(self, tz):
         result = to_datetime(arr)
         assert result is arr
 
+    @pytest.mark.parametrize(
+        "init_constructor, end_constructor",
+        [
+            (Index, DatetimeIndex),
+            (list, DatetimeIndex),
+            (np.array, DatetimeIndex),
+            (Series, Series),
+        ],
+    )
+    def test_to_datetime_arraylike_contains_pydatetime(
+        self, init_constructor, end_constructor
+    ):
+        # GH 49298
+        data = ["01/02/01 12:00", datetime(2001, 2, 2, 12, 30)]
+        expected_data = [
+            Timestamp("2001-02-01 12:00:00"),
+            Timestamp("2001-02-02 12:30:00"),
+        ]
+        result = to_datetime(init_constructor(data), format="%d/%m/%y %H:%M")
+        expected = end_constructor(expected_data)
+        tm.assert_equal(result, expected)
+
+    @pytest.mark.parametrize(
+        "init_constructor, end_constructor",
+        [
+            (Index, DatetimeIndex),
+            (list, DatetimeIndex),
+            (np.array, DatetimeIndex),
+            (Series, Series),
+        ],
+    )
+    def test_to_datetime_arraylike_contains_pydatetime_utc(
+        self, cache, init_constructor, end_constructor
+    ):
+        # GH 49298
+        dt = datetime(2010, 1, 2, 12, 13, 16)
+        dt = dt.replace(tzinfo=timezone.utc)
+        data = ["20100102 121314", "20100102 121315", dt]
+        expected_data = [
+            Timestamp("2010-01-02 12:13:14", tz="utc"),
+            Timestamp("2010-01-02 12:13:15", tz="utc"),
+            Timestamp("2010-01-02 12:13:16", tz="utc"),
+        ]
+
+        result = to_datetime(
+            init_constructor(data), format="%Y%m%d %H%M%S", utc=True, cache=cache
+        )
+        expected = end_constructor(expected_data)
+        tm.assert_equal(result, expected)
+
     def test_to_datetime_pydatetime(self):
         actual = to_datetime(datetime(2008, 1, 15))
         assert actual == datetime(2008, 1, 15)