asan checks for datetime funcs

pandas-dev · mroeschke · Dec 21, 2023 · Sep 11, 2023 · Sep 11, 2023 · Sep 12, 2023
commit 663d6d44c1aeb08e55fd4595217837f35dd3fe95
diff --git a/pandas/_libs/tslib.pyx b/pandas/_libs/tslib.pyx
@@ -94,8 +94,8 @@ def _test_parse_iso8601(ts: str):
     obj = _TSObject()
 
     string_to_dts(ts, &obj.dts, &out_bestunit, &out_local, &out_tzoffset, True)
+    check_dts_bounds(&obj.dts, NPY_FR_ns)
     obj.value = npy_datetimestruct_to_datetime(NPY_FR_ns, &obj.dts)
-    check_dts_bounds(&obj.dts)
     if out_local == 1:
         obj.tzinfo = timezone(timedelta(minutes=out_tzoffset))
         obj.value = tz_localize_to_utc_single(obj.value, obj.tzinfo)

diff --git a/pandas/_libs/tslibs/conversion.pyx b/pandas/_libs/tslibs/conversion.pyx
@@ -397,14 +397,14 @@ cdef _TSObject convert_datetime_to_tsobject(
     if nanos:
         obj.dts.ps = nanos * 1000
 
+    check_dts_bounds(&obj.dts, reso)
     obj.value = npy_datetimestruct_to_datetime(reso, &obj.dts)
 
     if obj.tzinfo is not None and not is_utc(obj.tzinfo):
         offset = get_utcoffset(obj.tzinfo, ts)
         pps = periods_per_second(reso)
         obj.value -= int(offset.total_seconds() * pps)
 
-    check_dts_bounds(&obj.dts, reso)
     check_overflows(obj, reso)
     return obj
 
@@ -434,6 +434,7 @@ cdef _TSObject _create_tsobject_tz_using_offset(npy_datetimestruct dts,
         datetime dt
         Py_ssize_t pos
 
+    check_dts_bounds(&dts, reso)
     value = npy_datetimestruct_to_datetime(reso, &dts)
     obj.dts = dts
     obj.tzinfo = timezone(timedelta(minutes=tzoffset))

diff --git a/pandas/_libs/tslibs/np_datetime.pyx b/pandas/_libs/tslibs/np_datetime.pyx
@@ -267,6 +267,7 @@ cdef int64_t pydatetime_to_dt64(datetime val,
     Note we are assuming that the datetime object is timezone-naive.
     """
     pydatetime_to_dtstruct(val, dts)
+    check_dts_bounds(dts, reso)
     return npy_datetimestruct_to_datetime(reso, dts)
 
 
@@ -282,6 +283,7 @@ cdef int64_t pydate_to_dt64(
     date val, npy_datetimestruct *dts, NPY_DATETIMEUNIT reso=NPY_FR_ns
 ):
     pydate_to_dtstruct(val, dts)
+    check_dts_bounds(dts, reso)
     return npy_datetimestruct_to_datetime(reso, dts)
 
 

diff --git a/pandas/_libs/tslibs/offsets.pyx b/pandas/_libs/tslibs/offsets.pyx
@@ -63,6 +63,7 @@ from pandas._libs.tslibs.nattype cimport (
 )
 from pandas._libs.tslibs.np_datetime cimport (
     NPY_DATETIMEUNIT,
+    check_dts_bounds,
     get_unit_from_dtype,
     import_pandas_datetime,
     npy_datetimestruct,
@@ -3093,6 +3094,8 @@ cdef class SemiMonthOffset(SingleConstructorOffset):
                     days_in_month = get_days_in_month(dts.year, dts.month)
                     dts.day = min(to_day, days_in_month)
 
+                    with gil:
+                        check_dts_bounds(&dts, reso)
                     res_val = npy_datetimestruct_to_datetime(reso, &dts)
 
                 # Analogous to: out[i] = res_val
@@ -4762,6 +4765,8 @@ cdef ndarray shift_quarters(
                 dts.month = month_add_months(dts, modby * n - months_since)
                 dts.day = get_day_of_month(&dts, day_opt)
 
+                with gil:
+                    check_dts_bounds(&dts, reso)
                 res_val = npy_datetimestruct_to_datetime(reso, &dts)
 
             # Analogous to: out[i] = res_val
@@ -4820,6 +4825,8 @@ def shift_months(
                     dts.month = month_add_months(dts, months)
 
                     dts.day = min(dts.day, get_days_in_month(dts.year, dts.month))
+                    with gil:
+                        check_dts_bounds(&dts, reso)
                     res_val = npy_datetimestruct_to_datetime(reso, &dts)
 
                 # Analogous to: out[i] = res_val
@@ -4846,6 +4853,8 @@ def shift_months(
                     dts.month = month_add_months(dts, months_to_roll)
                     dts.day = get_day_of_month(&dts, day_opt)
 
+                    with gil:
+                        check_dts_bounds(&dts, reso)
                     res_val = npy_datetimestruct_to_datetime(reso, &dts)
 
                 # Analogous to: out[i] = res_val

diff --git a/pandas/_libs/tslibs/period.pyx b/pandas/_libs/tslibs/period.pyx
@@ -721,6 +721,8 @@ cdef int64_t unix_date_from_ymd(int year, int month, int day) noexcept nogil:
         npy_datetimestruct dts
         int64_t unix_date
 
+    with gil:
+        check_dts_bounds(&dts, NPY_FR_D)
     memset(&dts, 0, sizeof(npy_datetimestruct))
     dts.year = year
     dts.month = month
@@ -738,6 +740,8 @@ cdef int64_t dts_to_year_ordinal(npy_datetimestruct *dts, int to_end) noexcept n
     cdef:
         int64_t result
 
+    with gil:
+        check_dts_bounds(dts, NPY_DATETIMEUNIT.NPY_FR_Y)
     result = npy_datetimestruct_to_datetime(NPY_DATETIMEUNIT.NPY_FR_Y, dts)
     if dts.month > to_end:
         return result + 1
@@ -795,14 +799,20 @@ cdef int64_t get_period_ordinal(npy_datetimestruct *dts, int freq) noexcept nogi
         return dts_to_qtr_ordinal(dts, fmonth)
 
     elif freq_group == FR_WK:
+        with gil:
+            check_dts_bounds(dts, NPY_FR_D)
         unix_date = npy_datetimestruct_to_datetime(NPY_FR_D, dts)
         return unix_date_to_week(unix_date, freq - FR_WK)
 
     elif freq == FR_BUS:
+        with gil:
+            check_dts_bounds(dts, NPY_FR_D)
         unix_date = npy_datetimestruct_to_datetime(NPY_FR_D, dts)
         return DtoB(dts, 0, unix_date)
 
     unit = freq_group_code_to_npy_unit(freq)
+    with gil:
+        check_dts_bounds(dts, unit)
     return npy_datetimestruct_to_datetime(unit, dts)
 
 

diff --git a/pandas/_libs/tslibs/strptime.pyx b/pandas/_libs/tslibs/strptime.pyx
@@ -325,6 +325,7 @@ def array_strptime(
             if string_to_dts_succeeded:
                 # No error reported by string_to_dts, pick back up
                    # where we left off
+                check_dts_bounds(&dts, NPY_FR_ns)
                 value = npy_datetimestruct_to_datetime(NPY_FR_ns, &dts)
                 if out_local == 1:
                     # Store the out_tzoffset in seconds
@@ -513,6 +514,7 @@ def array_strptime(
             dts.us = us
             dts.ps = ns * 1000
 
+            check_dts_bounds(&dts, NPY_FR_ns)
             iresult[i] = npy_datetimestruct_to_datetime(NPY_FR_ns, &dts)
             check_dts_bounds(&dts)
 

diff --git a/pandas/_libs/tslibs/timestamps.pyx b/pandas/_libs/tslibs/timestamps.pyx
@@ -581,6 +581,7 @@ cdef class _Timestamp(ABCTimestamp):
 
         if own_tz is not None and not is_utc(own_tz):
             pydatetime_to_dtstruct(self, &dts)
+            check_dts_bounds(&dts, self._creso)
             val = npy_datetimestruct_to_datetime(self._creso, &dts) + self.nanosecond
         else:
             val = self._value