BUG: Timestamp rounding wrong implementation fixed #11963

Author: Tux1

doc/source/api.rst

@@ -483,6 +483,8 @@ These can be accessed like ``Series.dt.<property>``.
    Series.dt.normalize
    Series.dt.strftime
    Series.dt.round
+   Series.dt.floor
+   Series.dt.ceil
 
 **Timedelta Properties**
 
@@ -1496,6 +1498,8 @@ Time-specific operations
    DatetimeIndex.tz_convert
    DatetimeIndex.tz_localize
    DatetimeIndex.round
+   DatetimeIndex.floor
+   DatetimeIndex.ceil
 
 Conversion
 ~~~~~~~~~~
@@ -1537,6 +1541,8 @@ Conversion
    TimedeltaIndex.to_pytimedelta
    TimedeltaIndex.to_series
    TimedeltaIndex.round
+   TimedeltaIndex.floor
+   TimedeltaIndex.ceil
 
 Window
 ------

doc/source/whatsnew/v0.18.0.txt

@@ -123,7 +123,7 @@ Other enhancements
 Datetimelike rounding
 ^^^^^^^^^^^^^^^^^^^^^
 
-``DatetimeIndex``, ``Timestamp``, ``TimedeltaIndex``, ``Timedelta`` have gained the ``.round()`` method for datetimelike rounding. (:issue:`4314`)
+``DatetimeIndex``, ``Timestamp``, ``TimedeltaIndex``, ``Timedelta`` have gained the ``.round()``, ``.floor()`` and ``.ceil()`` method for datetimelike rounding, flooring and ceiling. (:issue:`4314`, :issue:`11963`)
 
 Naive datetimes
 
@@ -137,7 +137,7 @@ Naive datetimes
    dr[0]
    dr[0].round('10s')
 
-Tz-aware are rounded in local times
+Tz-aware are rounded, floored and ceiled in local times
 
 .. ipython:: python
 
@@ -158,7 +158,7 @@ Timedeltas
    t[0].round('2h')
 
 
-In addition, ``.round()`` will be available thru the ``.dt`` accessor of ``Series``.
+In addition, ``.round()``, ``.floor()`` and ``.ceil()`` will be available thru the ``.dt`` accessor of ``Series``.
 
 .. ipython:: python
 

pandas/tests/test_categorical.py

@@ -3761,6 +3761,8 @@ def test_dt_accessor_api_for_categorical(self):
             ('strftime', ("%Y-%m-%d",), {}),
             ('tz_convert', ("EST",), {}),
             ('round', ("D",), {}),
+            ('floor', ("D",), {}),
+            ('ceil', ("D",), {}),
             #('tz_localize', ("UTC",), {}),
         ]
         _special_func_names = [f[0] for f in special_func_defs]

pandas/tests/test_series.py

@@ -89,9 +89,9 @@ def test_dt_namespace_accessor(self):
         ok_for_period_methods = ['strftime']
         ok_for_dt = ok_for_base + ['date','time','microsecond','nanosecond', 'is_month_start', 'is_month_end', 'is_quarter_start',
                                    'is_quarter_end', 'is_year_start', 'is_year_end', 'tz']
-        ok_for_dt_methods = ['to_period','to_pydatetime','tz_localize','tz_convert', 'normalize', 'strftime', 'round']
+        ok_for_dt_methods = ['to_period','to_pydatetime','tz_localize','tz_convert', 'normalize', 'strftime', 'round', 'floor', 'ceil']
         ok_for_td = ['days','seconds','microseconds','nanoseconds']
-        ok_for_td_methods = ['components','to_pytimedelta','total_seconds','round']
+        ok_for_td_methods = ['components','to_pytimedelta','total_seconds','round', 'floor', 'ceil']
 
         def get_expected(s, name):
             result = getattr(Index(s._values),prop)
@@ -141,14 +141,26 @@ def compare(s, name):
             tm.assert_series_equal(result, expected)
 
         # round
-        s = Series(date_range('20130101 09:10:11',periods=5))
+        s = Series(pd.to_datetime(['2012-01-01 13:00:00', '2012-01-01 12:01:00', '2012-01-01 08:00:00']))
         result = s.dt.round('D')
-        expected = Series(date_range('20130101',periods=5))
+        expected = Series(pd.to_datetime(['2012-01-02', '2012-01-02', '2012-01-01']))
         tm.assert_series_equal(result, expected)
 
         # round with tz
         result = s.dt.tz_localize('UTC').dt.tz_convert('US/Eastern').dt.round('D')
-        expected = Series(date_range('20130101',periods=5)).dt.tz_localize('US/Eastern')
+        expected = Series(pd.to_datetime(['2012-01-01', '2012-01-01', '2012-01-01']).tz_localize('US/Eastern'))
+        tm.assert_series_equal(result, expected)
+
+        # floor
+        s = Series(pd.to_datetime(['2012-01-01 13:00:00', '2012-01-01 12:01:00', '2012-01-01 08:00:00']))
+        result = s.dt.floor('D')
+        expected = Series(pd.to_datetime(['2012-01-01', '2012-01-01', '2012-01-01']))
+        tm.assert_series_equal(result, expected)
+
+        # ceil
+        s = Series(pd.to_datetime(['2012-01-01 13:00:00', '2012-01-01 12:01:00', '2012-01-01 08:00:00']))
+        result = s.dt.ceil('D')
+        expected = Series(pd.to_datetime(['2012-01-02', '2012-01-02', '2012-01-02']))
         tm.assert_series_equal(result, expected)
 
         # datetimeindex with tz

pandas/tseries/base.py

@@ -44,9 +44,9 @@ def strftime(self, date_format):
 class TimelikeOps(object):
     """ common ops for TimedeltaIndex/DatetimeIndex, but not PeriodIndex """
 
-    def round(self, freq):
+    _round_doc = (
         """
-        Round the index to the specified freq; this is a floor type of operation
+        %s the index to the specified freq
 
         Parameters
         ----------
@@ -59,7 +59,8 @@ def round(self, freq):
         Raises
         ------
         ValueError if the freq cannot be converted
-        """
+        """)
+    def _round(self, freq, rounder):
 
         from pandas.tseries.frequencies import to_offset
         unit = to_offset(freq).nanos
@@ -69,7 +70,7 @@ def round(self, freq):
             values = self.tz_localize(None).asi8
         else:
             values = self.asi8
-        result = (unit*np.floor(values/unit)).astype('i8')
+        result = (unit*rounder(values/float(unit))).astype('i8')
         attribs = self._get_attributes_dict()
         if 'freq' in attribs:
             attribs['freq'] = None
@@ -81,6 +82,18 @@ def round(self, freq):
         if getattr(self,'tz',None) is not None:
             result = result.tz_localize(self.tz)
         return result
+    
+    @Appender(_round_doc % "round")
+    def round(self, freq):
+       return self._round(freq, np.round)
+
+    @Appender(_round_doc % "floor")
+    def floor(self, freq):
+       return self._round(freq, np.floor)
+
+    @Appender(_round_doc % "floor")
+    def ceil(self, freq):
+       return self._round(freq, np.ceil)
 
 class DatetimeIndexOpsMixin(object):
     """ common ops mixin to support a unified inteface datetimelike Index """

pandas/tseries/common.py

@@ -146,7 +146,7 @@ def to_pydatetime(self):
                                            typ='property')
 DatetimeProperties._add_delegate_accessors(delegate=DatetimeIndex,
                                            accessors=["to_period","tz_localize","tz_convert",
-                                                      "normalize","strftime","round"],
+                                                      "normalize","strftime","round", "floor", "ceil"],
                                            typ='method')
 
 class TimedeltaProperties(Properties):
@@ -182,7 +182,7 @@ def components(self):
                                             accessors=TimedeltaIndex._datetimelike_ops,
                                             typ='property')
 TimedeltaProperties._add_delegate_accessors(delegate=TimedeltaIndex,
-                                            accessors=["to_pytimedelta", "total_seconds", "round"],
+                                            accessors=["to_pytimedelta", "total_seconds", "round", "floor", "ceil"],
                                             typ='method')
 
 class PeriodProperties(Properties):

pandas/tseries/tests/test_timedeltas.py

@@ -193,8 +193,8 @@ def test_round(self):
                                 Timedelta('-1 days 02:34:56.789000000')
                                 ),
                                ('S',
-                                Timedelta('1 days 02:34:56'),
-                                Timedelta('-1 days 02:34:56')
+                                Timedelta('1 days 02:34:57'),
+                                Timedelta('-1 days 02:34:57')
                                 ),
                                ('2S',
                                 Timedelta('1 days 02:34:56'),
@@ -205,15 +205,15 @@ def test_round(self):
                                 Timedelta('-1 days 02:34:55')
                                 ),
                                ('T',
-                                Timedelta('1 days 02:34:00'),
-                                Timedelta('-1 days 02:34:00')
+                                Timedelta('1 days 02:35:00'),
+                                Timedelta('-1 days 02:35:00')
                                 ),
                                ('12T',
-                                Timedelta('1 days 02:24:00'),
-                                Timedelta('-1 days 02:24:00')),
+                                Timedelta('1 days 02:36:00'),
+                                Timedelta('-1 days 02:36:00')),
                                ('H',
-                                Timedelta('1 days 02:00:00'),
-                                Timedelta('-1 days 02:00:00')
+                                Timedelta('1 days 03:00:00'),
+                                Timedelta('-1 days 03:00:00')
                                 ),
                                ('d',
                                 Timedelta('1 days'),
@@ -237,22 +237,15 @@ def test_round(self):
         # note that negative times round DOWN! so don't give whole numbers
         for (freq, s1, s2) in [('N', t1, t2),
                                ('U', t1, t2),
-                               ('L', t1a, TimedeltaIndex(['-1 days +00:00:00', '-2 days +23:58:57.999000',
-                                                          '-2 days +23:57:55.999000'],
+                               ('L', t1a, TimedeltaIndex(['-1 days +00:00:00', '-2 days +23:58:58', '-2 days +23:57:56'],
                                                          dtype='timedelta64[ns]', freq=None)),
-                               ('S', t1a, TimedeltaIndex(['-1 days +00:00:00', '-2 days +23:58:57', '-2 days +23:57:55'],
+                               ('S', t1a, TimedeltaIndex(['-1 days +00:00:00', '-2 days +23:58:58', '-2 days +23:57:56'],
                                                          dtype='timedelta64[ns]', freq=None)),
-                               ('2S', t1a, TimedeltaIndex(['-1 days +00:00:00', '-2 days +23:58:56', '-2 days +23:57:54'],
-                                                          dtype='timedelta64[ns]', freq=None)),
-                               ('5S', t1b, TimedeltaIndex(['-1 days +00:00:00', '-2 days +23:58:55', '-2 days +23:57:55'],
-                                                          dtype='timedelta64[ns]', freq=None)),
-                               ('T', t1b,  TimedeltaIndex(['-1 days +00:00:00', '-2 days +23:58:00', '-2 days +23:57:00'],
-                                                          dtype='timedelta64[ns]', freq=None)),
-                               ('12T', t1c, TimedeltaIndex(['-1 days +00:00:00', '-2 days +23:48:00', '-2 days +23:48:00'],
+                               ('12T', t1c, TimedeltaIndex(['-1 days', '-1 days', '-1 days'],
                                                            dtype='timedelta64[ns]', freq=None)),
-                               ('H',  t1c, TimedeltaIndex(['-1 days +00:00:00', '-2 days +23:00:00', '-2 days +23:00:00'],
+                               ('H',  t1c, TimedeltaIndex(['-1 days', '-1 days', '-1 days'],
                                                           dtype='timedelta64[ns]', freq=None)),
-                               ('d',  t1c, pd.TimedeltaIndex([-1,-2,-2],unit='D'))]:
+                               ('d',  t1c, pd.TimedeltaIndex([-1,-1,-1],unit='D'))]:
             r1 = t1.round(freq)
             tm.assert_index_equal(r1, s1)
             r2 = t2.round(freq)

pandas/tseries/tests/test_timeseries.py

@@ -2785,11 +2785,43 @@ def test_round(self):
         expected = Timestamp('20130101')
         self.assertEqual(result, expected)
 
+        dt = Timestamp('20130101 19:10:11')
+        result = dt.round('D')
+        expected = Timestamp('20130102')
+        self.assertEqual(result, expected)
+
+        dt = Timestamp('20130201 12:00:00')
+        result = dt.round('D')
+        expected = Timestamp('20130202')
+        self.assertEqual(result, expected)
+
+        dt = Timestamp('20130104 12:00:00')
+        result = dt.round('D')
+        expected = Timestamp('20130105')
+        self.assertEqual(result, expected)
+
+        dt = Timestamp('20130104 12:32:00')
+        result = dt.round('30Min')
+        expected = Timestamp('20130104 12:30:00')
+        self.assertEqual(result, expected)
+
         dti = date_range('20130101 09:10:11',periods=5)
         result = dti.round('D')
         expected = date_range('20130101',periods=5)
         tm.assert_index_equal(result, expected)
 
+        # floor
+        dt = Timestamp('20130101 09:10:11')
+        result = dt.floor('D')
+        expected = Timestamp('20130101')
+        self.assertEqual(result, expected)
+
+        # ceil
+        dt = Timestamp('20130101 09:10:11')
+        result = dt.ceil('D')
+        expected = Timestamp('20130102')
+        self.assertEqual(result, expected)
+
         # round with tz
         dt = Timestamp('20130101 09:10:11',tz='US/Eastern')
         result = dt.round('D')

pandas/tslib.pyx

@@ -361,14 +361,8 @@ class Timestamp(_Timestamp):
     def _repr_base(self):
         return '%s %s' % (self._date_repr, self._time_repr)
 
-    def round(self, freq):
-        """
-        return a new Timestamp rounded to this resolution
+    def _round(self, freq, rounder):
 
-        Parameters
-        ----------
-        freq : a freq string indicating the rouding resolution
-        """
         cdef int64_t unit
         cdef object result, value
 
@@ -378,11 +372,41 @@ class Timestamp(_Timestamp):
             value = self.tz_localize(None).value
         else:
             value = self.value
-        result = Timestamp(unit*np.floor(value/unit),unit='ns')
+        result = Timestamp(unit*rounder(value/float(unit)),unit='ns')
         if self.tz is not None:
             result = result.tz_localize(self.tz)
         return result
 
+    def round(self, freq):
+        """
+        return a new Timestamp rounded to this resolution
+
+        Parameters
+        ----------
+        freq : a freq string indicating the rounding resolution
+        """
+        return self._round(freq, np.round)
+
+    def floor(self, freq):
+        """
+        return a new Timestamp floored to this resolution
+
+        Parameters
+        ----------
+        freq : a freq string indicating the flooring resolution
+        """
+        return self._round(freq, np.floor)
+
+    def ceil(self, freq):
+        """
+        return a new Timestamp ceiled to this resolution
+
+        Parameters
+        ----------
+        freq : a freq string indicating the ceiling resolution
+        """
+        return self._round(freq, np.ceil)
+
     @property
     def tz(self):
         """
@@ -2388,20 +2412,45 @@ class Timedelta(_Timedelta):
         else:
            return "D"
 
+    def _round(self, freq, rounder):
+
+        cdef int64_t result, unit
+
+        from pandas.tseries.frequencies import to_offset
+        unit = to_offset(freq).nanos
+        result = unit*rounder(self.value/float(unit))
+        return Timedelta(result,unit='ns')
+
     def round(self, freq):
         """
-        return a new Timedelta rounded to this resolution
+        return a new Timedelta rounded to this resolution.
+
 
         Parameters
         ----------
-        freq : a freq string indicating the rouding resolution
+        freq : a freq string indicating the rounding resolution
         """
-        cdef int64_t result, unit
+        return self._round(freq, np.round)
 
-        from pandas.tseries.frequencies import to_offset
-        unit = to_offset(freq).nanos
-        result = unit*np.floor(self.value/unit)
-        return Timedelta(result,unit='ns')
+    def floor(self, freq):
+        """
+        return a new Timedelta floored to this resolution
+
+        Parameters
+        ----------
+        freq : a freq string indicating the flooring resolution
+        """
+        return self._round(freq, np.floor)
+
+    def ceil(self, freq):
+        """
+        return a new Timedelta ceiled to this resolution
+
+        Parameters
+        ----------
+        freq : a freq string indicating the ceiling resolution
+        """
+        return self._round(freq, np.ceil)
 
     def _repr_base(self, format=None):
         """