CLN: Move some PI/DTI methods to EA subclasses, implement tests (pandas-dev#22961)

Author: jbrockmendel

pandas/core/arrays/datetimes.py

@@ -31,7 +31,7 @@
 from pandas.core.algorithms import checked_add_with_arr
 from pandas.core import ops
 
-from pandas.tseries.frequencies import to_offset
+from pandas.tseries.frequencies import to_offset, get_period_alias
 from pandas.tseries.offsets import Tick, generate_range
 
 from pandas.core.arrays import datetimelike as dtl
@@ -200,6 +200,10 @@ def __new__(cls, values, freq=None, tz=None, dtype=None):
             # e.g. DatetimeIndex
             tz = values.tz
 
+        if freq is None and hasattr(values, "freq"):
+            # i.e. DatetimeArray, DatetimeIndex
+            freq = values.freq
+
         freq, freq_infer = dtl.maybe_infer_freq(freq)
 
         # if dtype has an embedded tz, capture it
@@ -764,6 +768,67 @@ def normalize(self):
         new_values = conversion.normalize_i8_timestamps(self.asi8, self.tz)
         return type(self)(new_values, freq='infer').tz_localize(self.tz)
 
+    def to_period(self, freq=None):
+        """
+        Cast to PeriodArray/Index at a particular frequency.
+
+        Converts DatetimeArray/Index to PeriodArray/Index.
+
+        Parameters
+        ----------
+        freq : string or Offset, optional
+            One of pandas' :ref:`offset strings <timeseries.offset_aliases>`
+            or an Offset object. Will be inferred by default.
+
+        Returns
+        -------
+        PeriodArray/Index
+
+        Raises
+        ------
+        ValueError
+            When converting a DatetimeArray/Index with non-regular values,
+            so that a frequency cannot be inferred.
+
+        Examples
+        --------
+        >>> df = pd.DataFrame({"y": [1,2,3]},
+        ...                   index=pd.to_datetime(["2000-03-31 00:00:00",
+        ...                                         "2000-05-31 00:00:00",
+        ...                                         "2000-08-31 00:00:00"]))
+        >>> df.index.to_period("M")
+        PeriodIndex(['2000-03', '2000-05', '2000-08'],
+                    dtype='period[M]', freq='M')
+
+        Infer the daily frequency
+
+        >>> idx = pd.date_range("2017-01-01", periods=2)
+        >>> idx.to_period()
+        PeriodIndex(['2017-01-01', '2017-01-02'],
+                    dtype='period[D]', freq='D')
+
+        See also
+        --------
+        pandas.PeriodIndex: Immutable ndarray holding ordinal values
+        pandas.DatetimeIndex.to_pydatetime: Return DatetimeIndex as object
+        """
+        from pandas.core.arrays.period import PeriodArrayMixin
+
+        if self.tz is not None:
+            warnings.warn("Converting to PeriodArray/Index representation "
+                          "will drop timezone information.", UserWarning)
+
+        if freq is None:
+            freq = self.freqstr or self.inferred_freq
+
+            if freq is None:
+                raise ValueError("You must pass a freq argument as "
+                                 "current index has none.")
+
+            freq = get_period_alias(freq)
+
+        return PeriodArrayMixin(self.values, freq=freq)
+
     # -----------------------------------------------------------------
     # Properties - Vectorized Timestamp Properties/Methods
 

pandas/core/arrays/period.py

@@ -10,14 +10,15 @@
     Period, IncompatibleFrequency, DIFFERENT_FREQ_INDEX,
     get_period_field_arr, period_asfreq_arr)
 from pandas._libs.tslibs import period as libperiod
-from pandas._libs.tslibs.timedeltas import delta_to_nanoseconds
+from pandas._libs.tslibs.timedeltas import delta_to_nanoseconds, Timedelta
 from pandas._libs.tslibs.fields import isleapyear_arr
 
 from pandas import compat
 from pandas.util._decorators import (cache_readonly, deprecate_kwarg)
 
 from pandas.core.dtypes.common import (
-    is_integer_dtype, is_float_dtype, is_period_dtype)
+    is_integer_dtype, is_float_dtype, is_period_dtype,
+    is_datetime64_dtype)
 from pandas.core.dtypes.dtypes import PeriodDtype
 from pandas.core.dtypes.generic import ABCSeries
 
@@ -127,6 +128,10 @@ def __new__(cls, values, freq=None, **kwargs):
             freq = values.freq
             values = values.asi8
 
+        elif is_datetime64_dtype(values):
+            # TODO: what if it has tz?
+            values = dt64arr_to_periodarr(values, freq)
+
         return cls._simple_new(values, freq, **kwargs)
 
     @classmethod
@@ -207,6 +212,14 @@ def is_leap_year(self):
         """ Logical indicating if the date belongs to a leap year """
         return isleapyear_arr(np.asarray(self.year))
 
+    @property
+    def start_time(self):
+        return self.to_timestamp(how='start')
+
+    @property
+    def end_time(self):
+        return self.to_timestamp(how='end')
+
     def asfreq(self, freq=None, how='E'):
         """
         Convert the Period Array/Index to the specified frequency `freq`.
@@ -266,6 +279,48 @@ def asfreq(self, freq=None, how='E'):
 
         return self._shallow_copy(new_data, freq=freq)
 
+    def to_timestamp(self, freq=None, how='start'):
+        """
+        Cast to DatetimeArray/Index
+
+        Parameters
+        ----------
+        freq : string or DateOffset, optional
+            Target frequency. The default is 'D' for week or longer,
+            'S' otherwise
+        how : {'s', 'e', 'start', 'end'}
+
+        Returns
+        -------
+        DatetimeArray/Index
+        """
+        from pandas.core.arrays.datetimes import DatetimeArrayMixin
+
+        how = libperiod._validate_end_alias(how)
+
+        end = how == 'E'
+        if end:
+            if freq == 'B':
+                # roll forward to ensure we land on B date
+                adjust = Timedelta(1, 'D') - Timedelta(1, 'ns')
+                return self.to_timestamp(how='start') + adjust
+            else:
+                adjust = Timedelta(1, 'ns')
+                return (self + 1).to_timestamp(how='start') - adjust
+
+        if freq is None:
+            base, mult = frequencies.get_freq_code(self.freq)
+            freq = frequencies.get_to_timestamp_base(base)
+        else:
+            freq = Period._maybe_convert_freq(freq)
+
+        base, mult = frequencies.get_freq_code(freq)
+        new_data = self.asfreq(freq, how=how)
+
+        new_data = libperiod.periodarr_to_dt64arr(new_data._ndarray_values,
+                                                  base)
+        return DatetimeArrayMixin(new_data, freq='infer')
+
     # ------------------------------------------------------------------
     # Arithmetic Methods
 
@@ -407,6 +462,15 @@ def _maybe_convert_timedelta(self, other):
 # -------------------------------------------------------------------
 # Constructor Helpers
 
+def dt64arr_to_periodarr(data, freq, tz=None):
+    if data.dtype != np.dtype('M8[ns]'):
+        raise ValueError('Wrong dtype: %s' % data.dtype)
+
+    freq = Period._maybe_convert_freq(freq)
+    base, mult = frequencies.get_freq_code(freq)
+    return libperiod.dt64arr_to_periodarr(data.view('i8'), base, tz)
+
+
 def _get_ordinal_range(start, end, periods, freq, mult=1):
     if com.count_not_none(start, end, periods) != 2:
         raise ValueError('Of the three parameters: start, end, and periods, '

pandas/core/indexes/datetimes.py

@@ -36,7 +36,7 @@
 from pandas.core.indexes.base import Index, _index_shared_docs
 from pandas.core.indexes.numeric import Int64Index, Float64Index
 import pandas.compat as compat
-from pandas.tseries.frequencies import to_offset, get_period_alias, Resolution
+from pandas.tseries.frequencies import to_offset, Resolution
 from pandas.core.indexes.datetimelike import (
     DatelikeOps, TimelikeOps, DatetimeIndexOpsMixin)
 from pandas.tseries.offsets import (
@@ -302,7 +302,8 @@ def __new__(cls, data=None,
                                        tz=tz, normalize=normalize,
                                        closed=closed, ambiguous=ambiguous)
 
-        if not isinstance(data, (np.ndarray, Index, ABCSeries)):
+        if not isinstance(data, (np.ndarray, Index, ABCSeries,
+                                 DatetimeArrayMixin)):
             if is_scalar(data):
                 raise ValueError('DatetimeIndex() must be called with a '
                                  'collection of some kind, %s was passed'
@@ -673,67 +674,12 @@ def to_series(self, keep_tz=False, index=None, name=None):
 
         return Series(values, index=index, name=name)
 
+    @Appender(DatetimeArrayMixin.to_period.__doc__)
     def to_period(self, freq=None):
-        """
-        Cast to PeriodIndex at a particular frequency.
-
-        Converts DatetimeIndex to PeriodIndex.
-
-        Parameters
-        ----------
-        freq : string or Offset, optional
-            One of pandas' :ref:`offset strings <timeseries.offset_aliases>`
-            or an Offset object. Will be inferred by default.
-
-        Returns
-        -------
-        PeriodIndex
-
-        Raises
-        ------
-        ValueError
-            When converting a DatetimeIndex with non-regular values, so that a
-            frequency cannot be inferred.
-
-        Examples
-        --------
-        >>> df = pd.DataFrame({"y": [1,2,3]},
-        ...                   index=pd.to_datetime(["2000-03-31 00:00:00",
-        ...                                         "2000-05-31 00:00:00",
-        ...                                         "2000-08-31 00:00:00"]))
-        >>> df.index.to_period("M")
-        PeriodIndex(['2000-03', '2000-05', '2000-08'],
-                    dtype='period[M]', freq='M')
-
-        Infer the daily frequency
-
-        >>> idx = pd.date_range("2017-01-01", periods=2)
-        >>> idx.to_period()
-        PeriodIndex(['2017-01-01', '2017-01-02'],
-                    dtype='period[D]', freq='D')
-
-        See also
-        --------
-        pandas.PeriodIndex: Immutable ndarray holding ordinal values
-        pandas.DatetimeIndex.to_pydatetime: Return DatetimeIndex as object
-        """
         from pandas.core.indexes.period import PeriodIndex
 
-        if self.tz is not None:
-            warnings.warn("Converting to PeriodIndex representation will "
-                          "drop timezone information.", UserWarning)
-
-        if freq is None:
-            freq = self.freqstr or self.inferred_freq
-
-            if freq is None:
-                msg = ("You must pass a freq argument as "
-                       "current index has none.")
-                raise ValueError(msg)
-
-            freq = get_period_alias(freq)
-
-        return PeriodIndex(self.values, name=self.name, freq=freq)
+        result = DatetimeArrayMixin.to_period(self, freq=freq)
+        return PeriodIndex(result, name=self.name)
 
     def snap(self, freq='S'):
         """
@@ -758,6 +704,7 @@ def snap(self, freq='S'):
 
         # we know it conforms; skip check
         return DatetimeIndex(snapped, freq=freq, verify_integrity=False)
+        # TODO: what about self.name?  if so, use shallow_copy?
 
     def unique(self, level=None):
         # Override here since IndexOpsMixin.unique uses self._values.unique
@@ -769,8 +716,7 @@ def unique(self, level=None):
         else:
             naive = self
         result = super(DatetimeIndex, naive).unique(level=level)
-        return self._simple_new(result.values, name=self.name, tz=self.tz,
-                                freq=self.freq)
+        return self._shallow_copy(result.values)
 
     def union(self, other):
         """
@@ -1421,8 +1367,7 @@ def insert(self, loc, item):
         try:
             new_dates = np.concatenate((self[:loc].asi8, [item.view(np.int64)],
                                         self[loc:].asi8))
-            return DatetimeIndex(new_dates, name=self.name, freq=freq,
-                                 tz=self.tz)
+            return self._shallow_copy(new_dates, freq=freq)
         except (AttributeError, TypeError):
 
             # fall back to object index
@@ -1458,7 +1403,7 @@ def delete(self, loc):
                 if (loc.start in (0, None) or loc.stop in (len(self), None)):
                     freq = self.freq
 
-        return DatetimeIndex(new_dates, name=self.name, freq=freq, tz=self.tz)
+        return self._shallow_copy(new_dates, freq=freq)
 
     def indexer_at_time(self, time, asof=False):
         """