BUG: Timestamp.replace handle out-of-pydatetime range (#50348)

* BUG: Timestamp.replace handle out-of-pydatetime range

* update test

* fix repr for out-of-pydatetime bounds

* update test

* xfail npdev

* Update pandas/_libs/tslibs/timestamps.pyx

Co-authored-by: Marco Edward Gorelli <[email protected]>

---------

Co-authored-by: Marco Edward Gorelli <[email protected]>

Author: jbrockmendel

pandas/_libs/tslibs/timestamps.pyx

@@ -8,6 +8,7 @@ shadows the python class, where we do any heavy lifting.
 """
 
 import warnings
+
 cimport cython
 
 import numpy as np
@@ -80,6 +81,7 @@ from pandas._libs.tslibs.nattype cimport (
 from pandas._libs.tslibs.np_datetime cimport (
     NPY_DATETIMEUNIT,
     NPY_FR_ns,
+    check_dts_bounds,
     cmp_dtstructs,
     cmp_scalar,
     convert_reso,
@@ -1032,19 +1034,19 @@ cdef class _Timestamp(ABCTimestamp):
         stamp = self._repr_base
         zone = None
 
-        try:
-            stamp += self.strftime("%z")
-        except ValueError:
-            year2000 = self.replace(year=2000)
-            stamp += year2000.strftime("%z")
+        if self.tzinfo is not None:
+            try:
+                stamp += self.strftime("%z")
+            except ValueError:
+                year2000 = self.replace(year=2000)
+                stamp += year2000.strftime("%z")
 
-        if self.tzinfo:
             zone = get_timezone(self.tzinfo)
-        try:
-            stamp += zone.strftime(" %%Z")
-        except AttributeError:
-            # e.g. tzlocal has no `strftime`
-            pass
+            try:
+                stamp += zone.strftime(" %%Z")
+            except AttributeError:
+                # e.g. tzlocal has no `strftime`
+                pass
 
         tz = f", tz='{zone}'" if zone is not None else ""
 
@@ -2216,6 +2218,7 @@ default 'raise'
             object k, v
             datetime ts_input
             tzinfo_type tzobj
+            _TSObject ts
 
         # set to naive if needed
         tzobj = self.tzinfo
@@ -2261,7 +2264,20 @@ default 'raise'
             tzobj = tzinfo
 
         # reconstruct & check bounds
-        if tzobj is not None and treat_tz_as_pytz(tzobj):
+        if tzobj is None:
+            # We can avoid going through pydatetime paths, which is robust
+            #  to datetimes outside of pydatetime range.
+            ts = _TSObject()
+            check_dts_bounds(&dts, self._creso)
+            ts.value = npy_datetimestruct_to_datetime(self._creso, &dts)
+            ts.dts = dts
+            ts.creso = self._creso
+            ts.fold = fold
+            return create_timestamp_from_ts(
+                ts.value, dts, tzobj, fold, reso=self._creso
+            )
+
+        elif tzobj is not None and treat_tz_as_pytz(tzobj):
             # replacing across a DST boundary may induce a new tzinfo object
             # see GH#18319
             ts_input = tzobj.localize(datetime(dts.year, dts.month, dts.day,

pandas/tests/scalar/timestamp/test_unary_ops.py

@@ -13,6 +13,7 @@
 from pandas._libs import lib
 from pandas._libs.tslibs import (
     NaT,
+    OutOfBoundsDatetime,
     Timedelta,
     Timestamp,
     conversion,
@@ -363,6 +364,19 @@ def checker(res, ts, nanos):
     # --------------------------------------------------------------
     # Timestamp.replace
 
+    def test_replace_out_of_pydatetime_bounds(self):
+        # GH#50348
+        ts = Timestamp("2016-01-01").as_unit("ns")
+
+        msg = "Out of bounds nanosecond timestamp: 99999-01-01 00:00:00"
+        with pytest.raises(OutOfBoundsDatetime, match=msg):
+            ts.replace(year=99_999)
+
+        ts = ts.as_unit("ms")
+        result = ts.replace(year=99_999)
+        assert result.year == 99_999
+        assert result._value == Timestamp(np.datetime64("99999-01-01", "ms"))._value
+
     def test_replace_non_nano(self):
         ts = Timestamp._from_value_and_reso(
             91514880000000000, NpyDatetimeUnit.NPY_FR_us.value, None

pandas/tests/tseries/offsets/test_year.py

@@ -7,8 +7,12 @@
 
 from datetime import datetime
 
+import numpy as np
 import pytest
 
+from pandas.compat import is_numpy_dev
+
+from pandas import Timestamp
 from pandas.tests.tseries.offsets.common import (
     assert_is_on_offset,
     assert_offset_equal,
@@ -317,3 +321,14 @@ def test_offset(self, case):
     def test_is_on_offset(self, case):
         offset, dt, expected = case
         assert_is_on_offset(offset, dt, expected)
+
+
+@pytest.mark.xfail(is_numpy_dev, reason="result year is 1973, unclear why")
+def test_add_out_of_pydatetime_range():
+    # GH#50348 don't raise in Timestamp.replace
+    ts = Timestamp(np.datetime64("-20000-12-31"))
+    off = YearEnd()
+
+    result = ts + off
+    expected = Timestamp(np.datetime64("-19999-12-31"))
+    assert result == expected