Merge pull request #11301 from jreback/tz_convert

BUG: Bug in tz-conversions with an ambiguous time and .dt accessors, #11295

Author: jreback

doc/source/whatsnew/v0.17.1.txt

@@ -55,8 +55,12 @@ Bug Fixes
 
 - Bug in ``.to_latex()`` output broken when the index has a name (:issue: `10660`)
 - Bug in ``HDFStore.append`` with strings whose encoded length exceded the max unencoded length (:issue:`11234`)
+
 - Bug in ``HDFStore.select`` when comparing with a numpy scalar in a where clause (:issue:`11283`)
 
+- Bug in tz-conversions with an ambiguous time and ``.dt`` accessors (:issues:`11295`)
+
+
 
 
 

pandas/tests/test_index.py

@@ -3348,6 +3348,15 @@ def test_construction_with_alt(self):
  i2 = DatetimeIndex(i.tz_localize(None).asi8, dtype=i.dtype, tz=i.dtype.tz)
  self.assert_index_equal(i, i2)
 
+ # localize into the provided tz
+ i2 = DatetimeIndex(i.tz_localize(None).asi8, tz='UTC')
+ expected = i.tz_localize(None).tz_localize('UTC')
+ self.assert_index_equal(i2, expected)
+
+ i2 = DatetimeIndex(i, tz='UTC')
+ expected = i.tz_convert('UTC')
+ self.assert_index_equal(i2, expected)
+
  # incompat tz/dtype
  self.assertRaises(ValueError, lambda : DatetimeIndex(i.tz_localize(None).asi8, dtype=i.dtype, tz='US/Pacific'))
 

pandas/tests/test_series.py

@@ -224,6 +224,18 @@ def get_dir(s):
  results = get_dir(s)
  tm.assert_almost_equal(results, list(sorted(set(ok_for_period + ok_for_period_methods))))
 
+ # 11295
+ # ambiguous time error on the conversions
+ s = Series(pd.date_range('2015-01-01', '2016-01-01', freq='T'))
+ s = s.dt.tz_localize('UTC').dt.tz_convert('America/Chicago')
+ results = get_dir(s)
+ tm.assert_almost_equal(results, list(sorted(set(ok_for_dt + ok_for_dt_methods))))
+ expected = Series(pd.date_range('2015-01-01',
+ '2016-01-01',
+ freq='T',
+ tz='UTC').tz_convert('America/Chicago'))
+ tm.assert_series_equal(s, expected)
+
  # no setting allowed
  s = Series(date_range('20130101',periods=5,freq='D'))
  with tm.assertRaisesRegexp(ValueError, "modifications"):

pandas/tseries/common.py

@@ -45,8 +45,10 @@ def maybe_to_datetimelike(data, copy=False):
  raise TypeError("cannot convert an object of type {0} to a datetimelike index".format(type(data)))
 
  index = data.index
- if is_datetime64_dtype(data.dtype) or is_datetime64tz_dtype(data.dtype):
+ if is_datetime64_dtype(data.dtype):
  return DatetimeProperties(DatetimeIndex(data, copy=copy, freq='infer'), index, name=data.name)
+ elif is_datetime64tz_dtype(data.dtype):
+ return DatetimeProperties(DatetimeIndex(data, copy=copy, freq='infer', ambiguous='infer'), index, name=data.name)
  elif is_timedelta64_dtype(data.dtype):
  return TimedeltaProperties(TimedeltaIndex(data, copy=copy, freq='infer'), index, name=data.name)
  else:

pandas/tseries/frequencies.py

@@ -851,10 +851,11 @@ def infer_freq(index, warn=True):
  raise TypeError("cannot infer freq from a non-convertible index type {0}".format(type(index)))
  index = index.values
 
- try:
- index = pd.DatetimeIndex(index)
- except AmbiguousTimeError:
- index = pd.DatetimeIndex(index.asi8)
+ if not isinstance(index, pd.DatetimeIndex):
+ try:
+ index = pd.DatetimeIndex(index)
+ except AmbiguousTimeError:
+ index = pd.DatetimeIndex(index.asi8)
 
  inferer = _FrequencyInferer(index, warn=warn)
  return inferer.get_freq()

pandas/tseries/index.py

@@ -194,7 +194,8 @@ def _join_i8_wrapper(joinf, **kwargs):
  _datetimelike_ops = ['year','month','day','hour','minute','second',
  'weekofyear','week','dayofweek','weekday','dayofyear','quarter', 'days_in_month', 'daysinmonth',
  'date','time','microsecond','nanosecond','is_month_start','is_month_end',
- 'is_quarter_start','is_quarter_end','is_year_start','is_year_end','tz','freq']
+ 'is_quarter_start','is_quarter_end','is_year_start','is_year_end',
+ 'tz','freq']
  _is_numeric_dtype = False
 
 
@@ -269,14 +270,7 @@ def __new__(cls, data=None,
  dayfirst=dayfirst,
  yearfirst=yearfirst)
 
- if is_datetimetz(data):
- # extract the data whether a Series or Index
- if isinstance(data, ABCSeries):
- data = data._values
- tz = data.tz
- data = data.tz_localize(None, ambiguous='infer').values
-
- if issubclass(data.dtype.type, np.datetime64):
+ if issubclass(data.dtype.type, np.datetime64) or is_datetimetz(data):
  if isinstance(data, ABCSeries):
  data = data._values
  if isinstance(data, DatetimeIndex):