From 490b189c2e8306709bdfab0c0bf145e0baf1832a Mon Sep 17 00:00:00 2001
From: Jon Wiggins <35540058+JonWiggins@users.noreply.github.com>
Date: Mon, 3 Jan 2022 18:25:23 -0600
Subject: [PATCH] BUG: Add fix for hashing timestamps with folds (#44282)

---
 doc/source/whatsnew/v1.4.0.rst                |  1 +
 pandas/_libs/tslibs/timestamps.pyx            |  2 +
 pandas/tests/frame/methods/test_reindex.py    | 37 +++++++++++++++++++
 .../tests/scalar/timestamp/test_timestamp.py  | 27 ++++++++++++++
 4 files changed, 67 insertions(+)

diff --git a/doc/source/whatsnew/v1.4.0.rst b/doc/source/whatsnew/v1.4.0.rst
index 029fb187740b3..6f724ec221c83 100644
--- a/doc/source/whatsnew/v1.4.0.rst
+++ b/doc/source/whatsnew/v1.4.0.rst
@@ -724,6 +724,7 @@ Datetimelike
 - Bug in :class:`DateOffset`` addition with :class:`Timestamp` where ``offset.nanoseconds`` would not be included in the result (:issue:`43968`, :issue:`36589`)
 - Bug in :meth:`Timestamp.fromtimestamp` not supporting the ``tz`` argument (:issue:`45083`)
 - Bug in :class:`DataFrame` construction from dict of :class:`Series` with mismatched index dtypes sometimes raising depending on the ordering of the passed dict (:issue:`44091`)
+- Bug in :class:`Timestamp` hashing during some DST transitions caused a segmentation fault (:issue:`33931` and :issue:`40817`)
 -
 
 Timedelta
diff --git a/pandas/_libs/tslibs/timestamps.pyx b/pandas/_libs/tslibs/timestamps.pyx
index 304ac9405c5e1..03ee62e59aa3d 100644
--- a/pandas/_libs/tslibs/timestamps.pyx
+++ b/pandas/_libs/tslibs/timestamps.pyx
@@ -180,6 +180,8 @@ cdef class _Timestamp(ABCTimestamp):
     def __hash__(_Timestamp self):
         if self.nanosecond:
             return hash(self.value)
+        if self.fold:
+            return datetime.__hash__(self.replace(fold=0))
         return datetime.__hash__(self)
 
     def __richcmp__(_Timestamp self, object other, int op):
diff --git a/pandas/tests/frame/methods/test_reindex.py b/pandas/tests/frame/methods/test_reindex.py
index bee8025275b42..b70ceea845ee8 100644
--- a/pandas/tests/frame/methods/test_reindex.py
+++ b/pandas/tests/frame/methods/test_reindex.py
@@ -8,6 +8,8 @@
 import numpy as np
 import pytest
 
+from pandas._libs.tslibs.timezones import dateutil_gettz as gettz
+
 import pandas as pd
 from pandas import (
     Categorical,
@@ -78,6 +80,41 @@ def test_setitem_reset_index_dtypes(self):
         result = df2.reset_index()
         tm.assert_frame_equal(result, expected)
 
+    @pytest.mark.parametrize(
+        "timezone, year, month, day, hour",
+        [["America/Chicago", 2013, 11, 3, 1], ["America/Santiago", 2021, 4, 3, 23]],
+    )
+    def test_reindex_timestamp_with_fold(self, timezone, year, month, day, hour):
+        # see gh-40817
+        test_timezone = gettz(timezone)
+        transition_1 = pd.Timestamp(
+            year=year,
+            month=month,
+            day=day,
+            hour=hour,
+            minute=0,
+            fold=0,
+            tzinfo=test_timezone,
+        )
+        transition_2 = pd.Timestamp(
+            year=year,
+            month=month,
+            day=day,
+            hour=hour,
+            minute=0,
+            fold=1,
+            tzinfo=test_timezone,
+        )
+        df = (
+            DataFrame({"index": [transition_1, transition_2], "vals": ["a", "b"]})
+            .set_index("index")
+            .reindex(["1", "2"])
+        )
+        tm.assert_frame_equal(
+            df,
+            DataFrame({"index": ["1", "2"], "vals": [None, None]}).set_index("index"),
+        )
+
 
 class TestDataFrameSelectReindex:
     # These are specific reindex-based tests; other indexing tests should go in
diff --git a/pandas/tests/scalar/timestamp/test_timestamp.py b/pandas/tests/scalar/timestamp/test_timestamp.py
index 1c34b5c8e5475..f734c331e3def 100644
--- a/pandas/tests/scalar/timestamp/test_timestamp.py
+++ b/pandas/tests/scalar/timestamp/test_timestamp.py
@@ -445,6 +445,33 @@ def test_hash_equivalent(self):
         stamp = Timestamp(datetime(2011, 1, 1))
         assert d[stamp] == 5
 
+    @pytest.mark.parametrize(
+        "timezone, year, month, day, hour",
+        [["America/Chicago", 2013, 11, 3, 1], ["America/Santiago", 2021, 4, 3, 23]],
+    )
+    def test_hash_timestamp_with_fold(self, timezone, year, month, day, hour):
+        # see gh-33931
+        test_timezone = gettz(timezone)
+        transition_1 = Timestamp(
+            year=year,
+            month=month,
+            day=day,
+            hour=hour,
+            minute=0,
+            fold=0,
+            tzinfo=test_timezone,
+        )
+        transition_2 = Timestamp(
+            year=year,
+            month=month,
+            day=day,
+            hour=hour,
+            minute=0,
+            fold=1,
+            tzinfo=test_timezone,
+        )
+        assert hash(transition_1) == hash(transition_2)
+
     def test_tz_conversion_freq(self, tz_naive_fixture):
         # GH25241
         with tm.assert_produces_warning(FutureWarning, match="freq"):