REF: de-duplicate tzinfo-awareness mismatch checks (#58171)

* REF: de-duplicate tzinfo-awareness mismatch checks

* function->method

Author: jbrockmendel

pandas/_libs/tslib.pyx

@@ -74,7 +74,6 @@ from pandas._libs.tslibs.nattype cimport (
     c_nat_strings as nat_strings,
 )
 from pandas._libs.tslibs.timestamps cimport _Timestamp
-from pandas._libs.tslibs.timezones cimport tz_compare
 
 from pandas._libs.tslibs import (
     Resolution,
@@ -452,13 +451,9 @@ cpdef array_to_datetime(
         ndarray[int64_t] iresult
         npy_datetimestruct dts
         bint utc_convert = bool(utc)
-        bint seen_datetime_offset = False
         bint is_raise = errors == "raise"
         bint is_coerce = errors == "coerce"
-        bint is_same_offsets
         _TSObject tsobj
-        float tz_offset
-        set out_tzoffset_vals = set()
         tzinfo tz, tz_out = None
         cnp.flatiter it = cnp.PyArray_IterNew(values)
         NPY_DATETIMEUNIT item_reso
@@ -568,12 +563,12 @@ cpdef array_to_datetime(
                     # dateutil timezone objects cannot be hashed, so
                     # store the UTC offsets in seconds instead
                     nsecs = tz.utcoffset(None).total_seconds()
-                    out_tzoffset_vals.add(nsecs)
-                    seen_datetime_offset = True
+                    state.out_tzoffset_vals.add(nsecs)
+                    state.found_aware_str = True
                 else:
                     # Add a marker for naive string, to track if we are
                     # parsing mixed naive and aware strings
-                    out_tzoffset_vals.add("naive")
+                    state.out_tzoffset_vals.add("naive")
                     state.found_naive_str = True
 
             else:
@@ -588,41 +583,7 @@ cpdef array_to_datetime(
                 raise
             return values, None
 
-    if seen_datetime_offset and not utc_convert:
-        # GH#17697, GH#57275
-        # 1) If all the offsets are equal, return one offset for
-        #    the parsed dates to (maybe) pass to DatetimeIndex
-        # 2) If the offsets are different, then do not force the parsing
-        #    and raise a ValueError: "cannot parse datetimes with
-        #    mixed time zones unless `utc=True`" instead
-        is_same_offsets = len(out_tzoffset_vals) == 1
-        if not is_same_offsets:
-            raise ValueError(
-                "Mixed timezones detected. Pass utc=True in to_datetime "
-                "or tz='UTC' in DatetimeIndex to convert to a common timezone."
-            )
-        elif state.found_naive or state.found_other:
-            # e.g. test_to_datetime_mixed_awareness_mixed_types
-            raise ValueError("Cannot mix tz-aware with tz-naive values")
-        elif tz_out is not None:
-            # GH#55693
-            tz_offset = out_tzoffset_vals.pop()
-            tz_out2 = timezone(timedelta(seconds=tz_offset))
-            if not tz_compare(tz_out, tz_out2):
-                # e.g. test_to_datetime_mixed_tzs_mixed_types
-                raise ValueError(
-                    "Mixed timezones detected. Pass utc=True in to_datetime "
-                    "or tz='UTC' in DatetimeIndex to convert to a common timezone."
-                )
-            # e.g. test_to_datetime_mixed_types_matching_tzs
-        else:
-            tz_offset = out_tzoffset_vals.pop()
-            tz_out = timezone(timedelta(seconds=tz_offset))
-    elif not utc_convert:
-        if tz_out and (state.found_other or state.found_naive_str):
-            # found_other indicates a tz-naive int, float, dt64, or date
-            # e.g. test_to_datetime_mixed_awareness_mixed_types
-            raise ValueError("Cannot mix tz-aware with tz-naive values")
+    tz_out = state.check_for_mixed_inputs(tz_out, utc)
 
     if infer_reso:
         if state.creso_ever_changed:

pandas/_libs/tslibs/strptime.pxd

@@ -18,9 +18,12 @@ cdef class DatetimeParseState:
         bint found_tz
         bint found_naive
         bint found_naive_str
+        bint found_aware_str
         bint found_other
         bint creso_ever_changed
         NPY_DATETIMEUNIT creso
+        set out_tzoffset_vals
 
     cdef tzinfo process_datetime(self, datetime dt, tzinfo tz, bint utc_convert)
     cdef bint update_creso(self, NPY_DATETIMEUNIT item_reso) noexcept
+    cdef tzinfo check_for_mixed_inputs(self, tzinfo tz_out, bint utc)

pandas/_libs/tslibs/strptime.pyx

@@ -252,8 +252,11 @@ cdef class DatetimeParseState:
         # found_naive_str refers to a string that was parsed to a timezone-naive
         #  datetime.
         self.found_naive_str = False
+        self.found_aware_str = False
         self.found_other = False
 
+        self.out_tzoffset_vals = set()
+
         self.creso = creso
         self.creso_ever_changed = False
 
@@ -292,6 +295,58 @@ cdef class DatetimeParseState:
                                  "tz-naive values")
         return tz
 
+    cdef tzinfo check_for_mixed_inputs(
+        self,
+        tzinfo tz_out,
+        bint utc,
+    ):
+        cdef:
+            bint is_same_offsets
+            float tz_offset
+
+        if self.found_aware_str and not utc:
+            # GH#17697, GH#57275
+            # 1) If all the offsets are equal, return one offset for
+            #    the parsed dates to (maybe) pass to DatetimeIndex
+            # 2) If the offsets are different, then do not force the parsing
+            #    and raise a ValueError: "cannot parse datetimes with
+            #    mixed time zones unless `utc=True`" instead
+            is_same_offsets = len(self.out_tzoffset_vals) == 1
+            if not is_same_offsets or (self.found_naive or self.found_other):
+                # e.g. test_to_datetime_mixed_awareness_mixed_types (array_to_datetime)
+                raise ValueError(
+                    "Mixed timezones detected. Pass utc=True in to_datetime "
+                    "or tz='UTC' in DatetimeIndex to convert to a common timezone."
+                )
+            elif tz_out is not None:
+                # GH#55693
+                tz_offset = self.out_tzoffset_vals.pop()
+                tz_out2 = timezone(timedelta(seconds=tz_offset))
+                if not tz_compare(tz_out, tz_out2):
+                    # e.g. (array_strptime)
+                    #  test_to_datetime_mixed_offsets_with_utc_false_removed
+                    # e.g. test_to_datetime_mixed_tzs_mixed_types (array_to_datetime)
+                    raise ValueError(
+                        "Mixed timezones detected. Pass utc=True in to_datetime "
+                        "or tz='UTC' in DatetimeIndex to convert to a common timezone."
+                    )
+                # e.g. (array_strptime)
+                #  test_guess_datetime_format_with_parseable_formats
+                # e.g. test_to_datetime_mixed_types_matching_tzs (array_to_datetime)
+            else:
+                # e.g. test_to_datetime_iso8601_with_timezone_valid (array_strptime)
+                tz_offset = self.out_tzoffset_vals.pop()
+                tz_out = timezone(timedelta(seconds=tz_offset))
+        elif not utc:
+            if tz_out and (self.found_other or self.found_naive_str):
+                # found_other indicates a tz-naive int, float, dt64, or date
+                # e.g. test_to_datetime_mixed_awareness_mixed_types (array_to_datetime)
+                raise ValueError(
+                    "Mixed timezones detected. Pass utc=True in to_datetime "
+                    "or tz='UTC' in DatetimeIndex to convert to a common timezone."
+                )
+        return tz_out
+
 
 def array_strptime(
     ndarray[object] values,
@@ -319,11 +374,8 @@ def array_strptime(
         npy_datetimestruct dts
         int64_t[::1] iresult
         object val
-        bint seen_datetime_offset = False
         bint is_raise = errors=="raise"
         bint is_coerce = errors=="coerce"
-        bint is_same_offsets
-        set out_tzoffset_vals = set()
         tzinfo tz, tz_out = None
         bint iso_format = format_is_iso(fmt)
         NPY_DATETIMEUNIT out_bestunit, item_reso
@@ -418,15 +470,15 @@ def array_strptime(
                     ) from err
                 if out_local == 1:
                     nsecs = out_tzoffset * 60
-                    out_tzoffset_vals.add(nsecs)
-                    seen_datetime_offset = True
+                    state.out_tzoffset_vals.add(nsecs)
+                    state.found_aware_str = True
                     tz = timezone(timedelta(minutes=out_tzoffset))
                     value = tz_localize_to_utc_single(
                         value, tz, ambiguous="raise", nonexistent=None, creso=creso
                     )
                 else:
                     tz = None
-                    out_tzoffset_vals.add("naive")
+                    state.out_tzoffset_vals.add("naive")
                     state.found_naive_str = True
                 iresult[i] = value
                 continue
@@ -475,12 +527,12 @@ def array_strptime(
                 elif creso == NPY_DATETIMEUNIT.NPY_FR_ms:
                     nsecs = nsecs // 10**3
 
-                out_tzoffset_vals.add(nsecs)
-                seen_datetime_offset = True
+                state.out_tzoffset_vals.add(nsecs)
+                state.found_aware_str = True
             else:
                 state.found_naive_str = True
                 tz = None
-                out_tzoffset_vals.add("naive")
+                state.out_tzoffset_vals.add("naive")
 
         except ValueError as ex:
             ex.args = (
@@ -499,35 +551,7 @@ def array_strptime(
                 raise
             return values, None
 
-    if seen_datetime_offset and not utc:
-        is_same_offsets = len(out_tzoffset_vals) == 1
-        if not is_same_offsets or (state.found_naive or state.found_other):
-            raise ValueError(
-                "Mixed timezones detected. Pass utc=True in to_datetime "
-                "or tz='UTC' in DatetimeIndex to convert to a common timezone."
-            )
-        elif tz_out is not None:
-            # GH#55693
-            tz_offset = out_tzoffset_vals.pop()
-            tz_out2 = timezone(timedelta(seconds=tz_offset))
-            if not tz_compare(tz_out, tz_out2):
-                # e.g. test_to_datetime_mixed_offsets_with_utc_false_removed
-                raise ValueError(
-                    "Mixed timezones detected. Pass utc=True in to_datetime "
-                    "or tz='UTC' in DatetimeIndex to convert to a common timezone."
-                )
-            # e.g. test_guess_datetime_format_with_parseable_formats
-        else:
-            # e.g. test_to_datetime_iso8601_with_timezone_valid
-            tz_offset = out_tzoffset_vals.pop()
-            tz_out = timezone(timedelta(seconds=tz_offset))
-    elif not utc:
-        if tz_out and (state.found_other or state.found_naive_str):
-            # found_other indicates a tz-naive int, float, dt64, or date
-            raise ValueError(
-                "Mixed timezones detected. Pass utc=True in to_datetime "
-                "or tz='UTC' in DatetimeIndex to convert to a common timezone."
-            )
+    tz_out = state.check_for_mixed_inputs(tz_out, utc)
 
     if infer_reso:
         if state.creso_ever_changed:

pandas/tests/tools/test_to_datetime.py

@@ -3545,19 +3545,27 @@ def test_to_datetime_mixed_awareness_mixed_types(aware_val, naive_val, naive_fir
     #  issued in _array_to_datetime_object
     both_strs = isinstance(aware_val, str) and isinstance(naive_val, str)
     has_numeric = isinstance(naive_val, (int, float))
+    both_datetime = isinstance(naive_val, datetime) and isinstance(aware_val, datetime)
+
+    mixed_msg = (
+        "Mixed timezones detected. Pass utc=True in to_datetime or tz='UTC' "
+        "in DatetimeIndex to convert to a common timezone"
+    )
 
     first_non_null = next(x for x in vec if x != "")
     # if first_non_null is a not a string, _guess_datetime_format_for_array
     #  doesn't guess a format so we don't go through array_strptime
     if not isinstance(first_non_null, str):
         # that case goes through array_strptime which has different behavior
-        msg = "Cannot mix tz-aware with tz-naive values"
+        msg = mixed_msg
         if naive_first and isinstance(aware_val, Timestamp):
             if isinstance(naive_val, Timestamp):
                 msg = "Tz-aware datetime.datetime cannot be converted to datetime64"
             with pytest.raises(ValueError, match=msg):
                 to_datetime(vec)
         else:
+            if not naive_first and both_datetime:
+                msg = "Cannot mix tz-aware with tz-naive values"
             with pytest.raises(ValueError, match=msg):
                 to_datetime(vec)
 
@@ -3586,21 +3594,21 @@ def test_to_datetime_mixed_awareness_mixed_types(aware_val, naive_val, naive_fir
             to_datetime(vec, utc=True)
 
     else:
-        msg = "Mixed timezones detected. Pass utc=True in to_datetime"
+        msg = mixed_msg
         with pytest.raises(ValueError, match=msg):
             to_datetime(vec)
 
         # No warning/error with utc=True
         to_datetime(vec, utc=True)
 
     if both_strs:
-        msg = "Mixed timezones detected. Pass utc=True in to_datetime"
+        msg = mixed_msg
         with pytest.raises(ValueError, match=msg):
             to_datetime(vec, format="mixed")
         with pytest.raises(ValueError, match=msg):
             DatetimeIndex(vec)
     else:
-        msg = "Cannot mix tz-aware with tz-naive values"
+        msg = mixed_msg
         if naive_first and isinstance(aware_val, Timestamp):
             if isinstance(naive_val, Timestamp):
                 msg = "Tz-aware datetime.datetime cannot be converted to datetime64"
@@ -3609,6 +3617,8 @@ def test_to_datetime_mixed_awareness_mixed_types(aware_val, naive_val, naive_fir
             with pytest.raises(ValueError, match=msg):
                 DatetimeIndex(vec)
         else:
+            if not naive_first and both_datetime:
+                msg = "Cannot mix tz-aware with tz-naive values"
             with pytest.raises(ValueError, match=msg):
                 to_datetime(vec, format="mixed")
             with pytest.raises(ValueError, match=msg):