Merge branch 'main' into fix/read_csv-large-num

Author: Alvaro-Kothe

doc/source/whatsnew/v3.0.0.rst

@@ -981,7 +981,8 @@ Timezones
 ^^^^^^^^^
 - Bug in :meth:`DatetimeIndex.union`, :meth:`DatetimeIndex.intersection`, and :meth:`DatetimeIndex.symmetric_difference` changing timezone to UTC when merging two DatetimeIndex objects with the same timezone but different units (:issue:`60080`)
 - Bug in :meth:`Series.dt.tz_localize` with a timezone-aware :class:`ArrowDtype` incorrectly converting to UTC when ``tz=None`` (:issue:`61780`)
--
+- Fixed bug in :func:`date_range` where tz-aware endpoints with calendar offsets (e.g. ``"MS"``) failed on DST fall-back. These now respect ``ambiguous``/ ``nonexistent``. (:issue:`52908`)
+
 
 Numeric
 ^^^^^^^

pandas/_libs/index.pyx

@@ -838,7 +838,7 @@ cdef class BaseMultiIndexCodesEngine:
             raise KeyError(key)
         try:
             indices = [1 if checknull(v) else lev.get_loc(v) + multiindex_nulls_shift
-                       for lev, v in zip(self.levels, key)]
+                       for lev, v in zip(self.levels, key, strict=True)]
         except KeyError:
             raise KeyError(key)
 

pandas/_libs/missing.pyx

@@ -72,7 +72,7 @@ cpdef bint check_na_tuples_nonequal(object left, object right):
     if len(left) != len(right):
         return False
 
-    for left_element, right_element in zip(left, right):
+    for left_element, right_element in zip(left, right, strict=True):
         if left_element is C_NA and right_element is not C_NA:
             return True
         elif right_element is C_NA and left_element is not C_NA:

pandas/_libs/tslibs/fields.pyx

@@ -109,7 +109,7 @@ def month_position_check(fields, weekdays) -> str | None:
         int32_t[:] months = fields["M"]
         int32_t[:] days = fields["D"]
 
-    for y, m, d, wd in zip(years, months, days, weekdays):
+    for y, m, d, wd in zip(years, months, days, weekdays, strict=True):
         if calendar_start:
             calendar_start &= d == 1
         if business_start:

pandas/_libs/tslibs/offsets.pyx

@@ -2217,7 +2217,7 @@ cdef class BusinessHour(BusinessMixin):
         # Use python string formatting to be faster than strftime
         hours = ",".join(
             f"{st.hour:02d}:{st.minute:02d}-{en.hour:02d}:{en.minute:02d}"
-            for st, en in zip(self.start, self.end)
+            for st, en in zip(self.start, self.end, strict=True)
         )
         attrs = [f"{self._prefix}={hours}"]
         out += ": " + ", ".join(attrs)
@@ -2414,7 +2414,7 @@ cdef class BusinessHour(BusinessMixin):
         # get total business hours by sec in one business day
         businesshours = sum(
             self._get_business_hours_by_sec(st, en)
-            for st, en in zip(self.start, self.end)
+            for st, en in zip(self.start, self.end, strict=True)
         )
 
         bd, r = divmod(abs(n * 60), businesshours // 60)
@@ -5357,7 +5357,7 @@ cpdef to_offset(freq, bint is_period=False):
                 # the last element must be blank
                 raise ValueError("last element must be blank")
 
-            tups = zip(split[0::4], split[1::4], split[2::4])
+            tups = zip(split[0::4], split[1::4], split[2::4], strict=False)
             for n, (sep, stride, name) in enumerate(tups):
                 name = _warn_about_deprecated_aliases(name, is_period)
                 _validate_to_offset_alias(name, is_period)

pandas/_libs/tslibs/timezones.pyx

@@ -252,7 +252,7 @@ cdef object _get_utc_trans_times_from_dateutil_tz(tzinfo tz):
     """
     new_trans = list(tz._trans_list)
     last_std_offset = 0
-    for i, (trans, tti) in enumerate(zip(tz._trans_list, tz._trans_idx)):
+    for i, (trans, tti) in enumerate(zip(tz._trans_list, tz._trans_idx, strict=True)):
         if not tti.isdst:
             last_std_offset = tti.offset
         new_trans[i] = trans - last_std_offset

pandas/core/arrays/datetimes.py

@@ -456,13 +456,14 @@ def _generate_range(
             end = _maybe_localize_point(end, freq, tz, ambiguous, nonexistent)
 
         if freq is not None:
-            # We break Day arithmetic (fixed 24 hour) here and opt for
-            # Day to mean calendar day (23/24/25 hour). Therefore, strip
-            # tz info from start and day to avoid DST arithmetic
-            if isinstance(freq, Day):
-                if start is not None:
+            # Offset handling:
+            # Ticks (fixed-duration like hours/minutes): keep tz; do absolute-time math.
+            # Other calendar offsets: drop tz; do naive wall time; localize once later
+            # so `ambiguous`/`nonexistent` are applied correctly.
+            if not isinstance(freq, Tick):
+                if start is not None and start.tz is not None:
                     start = start.tz_localize(None)
-                if end is not None:
+                if end is not None and end.tz is not None:
                     end = end.tz_localize(None)
 
             if isinstance(freq, (Tick, Day)):

pandas/core/frame.py

@@ -107,6 +107,7 @@
     is_list_like,
     is_scalar,
     is_sequence,
+    is_string_dtype,
     needs_i8_conversion,
     pandas_dtype,
 )
@@ -4454,8 +4455,12 @@ def _set_item_frame_value(self, key, value: DataFrame) -> None:
                 cols_droplevel = maybe_droplevels(cols, key)
                 if (
                     not isinstance(cols_droplevel, MultiIndex)
+                    and is_string_dtype(cols_droplevel.dtype)
                     and not cols_droplevel.any()
                 ):
+                    # if cols_droplevel contains only empty strings,
+                    # value.reindex(cols_droplevel, axis=1) would be full of NaNs
+                    # see GH#62518 and GH#61841
                     return
                 if len(cols_droplevel) and not cols_droplevel.equals(value.columns):
                     value = value.reindex(cols_droplevel, axis=1)

pandas/core/window/rolling.py

@@ -351,7 +351,7 @@ def __iter__(self) -> Iterator:
         )
         self._check_window_bounds(start, end, len(obj))
 
-        for s, e in zip(start, end):
+        for s, e in zip(start, end, strict=True):
             result = obj.iloc[slice(s, e)]
             yield result
 
@@ -802,7 +802,7 @@ def _apply_pairwise(
             groupby_codes = []
             groupby_levels = []
             # e.g. [[1, 2], [4, 5]] as [[1, 4], [2, 5]]
-            for gb_level_pair in map(list, zip(*gb_pairs)):
+            for gb_level_pair in map(list, zip(*gb_pairs, strict=True)):
                 labels = np.repeat(np.array(gb_level_pair), old_result_len)
                 codes, levels = factorize(labels)
                 groupby_codes.append(codes)