Separate out _convert_datetime_to_tsobject

pandas/_libs/tslib.pyx

@@ -728,7 +728,7 @@ class Timestamp(_Timestamp):
  # reconstruct & check bounds
  ts_input = datetime(dts.year, dts.month, dts.day, dts.hour, dts.min,
  dts.sec, dts.us, tzinfo=_tzinfo)
- ts = convert_to_tsobject(ts_input, _tzinfo, None, 0, 0)
+ ts = convert_datetime_to_tsobject(ts_input, _tzinfo)
  value = ts.value + (dts.ps // 1000)
  if value != NPY_NAT:
  _check_dts_bounds(&dts)
@@ -1475,52 +1475,11 @@ cdef convert_to_tsobject(object ts, object tz, object unit,
  obj.value = ts
  pandas_datetime_to_datetimestruct(ts, PANDAS_FR_ns, &obj.dts)
  elif PyDateTime_Check(ts):
- if tz is not None:
- # sort of a temporary hack
- if ts.tzinfo is not None:
- if (hasattr(tz, 'normalize') and
- hasattr(ts.tzinfo, '_utcoffset')):
- ts = tz.normalize(ts)
- obj.value = _pydatetime_to_dts(ts, &obj.dts)
- obj.tzinfo = ts.tzinfo
- else: #tzoffset
- try:
- tz = ts.astimezone(tz).tzinfo
- except:
- pass
- obj.value = _pydatetime_to_dts(ts, &obj.dts)
- ts_offset = get_utcoffset(ts.tzinfo, ts)
- obj.value -= _delta_to_nanoseconds(ts_offset)
- tz_offset = get_utcoffset(tz, ts)
- obj.value += _delta_to_nanoseconds(tz_offset)
- pandas_datetime_to_datetimestruct(obj.value,
- PANDAS_FR_ns, &obj.dts)
- obj.tzinfo = tz
- elif not is_utc(tz):
- ts = _localize_pydatetime(ts, tz)
- obj.value = _pydatetime_to_dts(ts, &obj.dts)
- obj.tzinfo = ts.tzinfo
- else:
- # UTC
- obj.value = _pydatetime_to_dts(ts, &obj.dts)
- obj.tzinfo = pytz.utc
- else:
- obj.value = _pydatetime_to_dts(ts, &obj.dts)
- obj.tzinfo = ts.tzinfo
-
- if obj.tzinfo is not None and not is_utc(obj.tzinfo):
- offset = get_utcoffset(obj.tzinfo, ts)
- obj.value -= _delta_to_nanoseconds(offset)
-
- if is_timestamp(ts):
- obj.value += ts.nanosecond
- obj.dts.ps = ts.nanosecond * 1000
- _check_dts_bounds(&obj.dts)
- return obj
+ return convert_datetime_to_tsobject(ts, tz)
  elif PyDate_Check(ts):
  # Keep the converter same as PyDateTime's
  ts = datetime.combine(ts, datetime_time())
- return convert_to_tsobject(ts, tz, None, 0, 0)
+ return convert_datetime_to_tsobject(ts, tz)
  elif getattr(ts, '_typ', None) == 'period':
  raise ValueError(
  "Cannot convert Period to Timestamp "
@@ -1538,6 +1497,83 @@ cdef convert_to_tsobject(object ts, object tz, object unit,
  return obj
 
 
+cdef _TSObject convert_datetime_to_tsobject(datetime ts, object tz,
+ int32_t nanos=0):
+ """
+ Convert a datetime (or Timestamp) input `ts`, along with optional timezone
+ object `tz` to a _TSObject.
+
+ The optional argument `nanos` allows for cases where datetime input
+ needs to be supplemented with higher-precision information.
+
+ Parameters
+ ----------
+ ts : datetime or Timestamp
+ Value to be converted to _TSObject
+ tz : tzinfo or None
+ timezone for the timezone-aware output
+ nanos : int32_t, default is 0
+ nanoseconds supplement the precision of the datetime input ts
+
+ Returns
+ -------
+ obj : _TSObject
+ """
+ cdef:
+ _TSObject obj = _TSObject()
+
+ if tz is not None:
+ tz = maybe_get_tz(tz)
+
+ # sort of a temporary hack
+ if ts.tzinfo is not None:
+ if (hasattr(tz, 'normalize') and
+ hasattr(ts.tzinfo, '_utcoffset')):
+ ts = tz.normalize(ts)
+ obj.value = _pydatetime_to_dts(ts, &obj.dts)
+ obj.tzinfo = ts.tzinfo
+ else:
+ # tzoffset
+ try:
+ tz = ts.astimezone(tz).tzinfo
+ except:
+ pass
+ obj.value = _pydatetime_to_dts(ts, &obj.dts)
+ ts_offset = get_utcoffset(ts.tzinfo, ts)
+ obj.value -= int(ts_offset.total_seconds() * 1e9)
+ tz_offset = get_utcoffset(tz, ts)
+ obj.value += int(tz_offset.total_seconds() * 1e9)
+ pandas_datetime_to_datetimestruct(obj.value,
+ PANDAS_FR_ns, &obj.dts)
+ obj.tzinfo = tz
+ elif not is_utc(tz):
+ ts = _localize_pydatetime(ts, tz)
+ obj.value = _pydatetime_to_dts(ts, &obj.dts)
+ obj.tzinfo = ts.tzinfo
+ else:
+ # UTC
+ obj.value = _pydatetime_to_dts(ts, &obj.dts)
+ obj.tzinfo = pytz.utc
+ else:
+ obj.value = _pydatetime_to_dts(ts, &obj.dts)
+ obj.tzinfo = ts.tzinfo
+
+ if obj.tzinfo is not None and not is_utc(obj.tzinfo):
+ offset = get_utcoffset(obj.tzinfo, ts)
+ obj.value -= int(offset.total_seconds() * 1e9)
+
+ if is_timestamp(ts):
+ obj.value += ts.nanosecond
+ obj.dts.ps = ts.nanosecond * 1000
+
+ if nanos:
+ obj.value += nanos
+ obj.dts.ps = nanos * 1000
+
+ _check_dts_bounds(&obj.dts)
+ return obj
+
+
 cpdef convert_str_to_tsobject(object ts, object tz, object unit,
  dayfirst=False, yearfirst=False):
  """ ts must be a string """
@@ -1558,11 +1594,12 @@ cpdef convert_str_to_tsobject(object ts, object tz, object unit,
  elif ts == 'now':
  # Issue 9000, we short-circuit rather than going
  # into np_datetime_strings which returns utc
- ts = Timestamp.now(tz)
+ ts = datetime.now(tz)
  elif ts == 'today':
  # Issue 9000, we short-circuit rather than going
  # into np_datetime_strings which returns a normalized datetime
- ts = Timestamp.today(tz)
+ ts = datetime.now(tz)
+ # equiv: datetime.today().replace(tzinfo=tz)
  else:
  try:
  _string_to_dts(ts, &obj.dts, &out_local, &out_tzoffset)
@@ -1576,8 +1613,8 @@ cpdef convert_str_to_tsobject(object ts, object tz, object unit,
  _check_dts_bounds(&obj.dts)
  return obj
  else:
- # Keep the converter same as PyDateTime's
- ts = Timestamp(obj.value, tz=obj.tzinfo)
+ return _convert_tsobject_tz(obj, tz)
+
  else:
  ts = obj.value
  if tz is not None:
@@ -1595,6 +1632,32 @@ cpdef convert_str_to_tsobject(object ts, object tz, object unit,
  return convert_to_tsobject(ts, tz, unit, dayfirst, yearfirst)
 
 
+cdef _TSObject _convert_tsobject_tz(_TSObject obj, object tz):
+ """
+ Given a _TSObject with a FixedOffset tzinfo, convert to a new _TSObject
+ instance with the given timezone `tz` (or tz-naive if tz is None).
+
+ Parameters
+ ----------
+ obj : _TSObject
+ tz : tzinfo or None
+
+ Returns
+ -------
+ obj : _TSObject
+ """
+ # Keep the converter same as PyDateTime's
+ cdef:
+ datetime dtime
+
+ obj = convert_to_tsobject(obj.value, obj.tzinfo, None, 0, 0)
+ dtime = datetime(obj.dts.year, obj.dts.month, obj.dts.day,
+ obj.dts.hour, obj.dts.min, obj.dts.sec, obj.dts.us,
+ tz)
+ obj = convert_datetime_to_tsobject(dtime, tz, nanos=obj.dts.ps / 1000)
+ return obj
+
+
 def _test_parse_iso8601(object ts):
  """
  TESTING ONLY: Parse string into Timestamp using iso8601 parser. Used
@@ -1726,7 +1789,7 @@ def datetime_to_datetime64(ndarray[object] values):
  else:
  inferred_tz = get_timezone(val.tzinfo)
 
- _ts = convert_to_tsobject(val, None, None, 0, 0)
+ _ts = convert_datetime_to_tsobject(val, None)
  iresult[i] = _ts.value
  _check_dts_bounds(&_ts.dts)
  else:
@@ -2046,7 +2109,7 @@ cpdef array_to_datetime(ndarray[object] values, errors='raise',
  seen_datetime=1
  if val.tzinfo is not None:
  if utc_convert:
- _ts = convert_to_tsobject(val, None, 'ns', 0, 0)
+ _ts = convert_datetime_to_tsobject(val, None)
  iresult[i] = _ts.value
  try:
  _check_dts_bounds(&_ts.dts)
@@ -2155,7 +2218,7 @@ cpdef array_to_datetime(ndarray[object] values, errors='raise',
  raise TypeError("invalid string coercion to datetime")
 
  try:
- _ts = convert_to_tsobject(py_dt, None, None, 0, 0)
+ _ts = convert_datetime_to_tsobject(py_dt, None)
  iresult[i] = _ts.value
  except ValueError:
  if is_coerce: