DEPR: convert_datetime64 parameter in to_records()

Author: reidy-p

doc/source/whatsnew/v0.23.0.txt

@@ -740,6 +740,7 @@ Deprecations
 
 - ``pandas.tseries.plotting.tsplot`` is deprecated. Use :func:`Series.plot` instead (:issue:`18627`)
 - ``Index.summary()`` is deprecated and will be removed in a future version (:issue:`18217`)
+- The ``convert_datetime64`` parameter has been deprecated and the default value is now ``False`` in :func:`to_records` as the NumPy bug motivating this parameter has been resolved (:issue:`18160`).
 
 .. _whatsnew_0230.prior_deprecations:
 

hack.py

@@ -0,0 +1,95 @@
+import pandas as pd 
+import pdb
+
+p = pd.Period('2017-01-01', 'D')
+pi = pd.PeriodIndex([p])
+
+#print(pd.Series(p).dt.end_time[0])
+#print(pd.Series(pi).dt.end_time[0])
+#
+#print(p.end_time)
+#print(pi.end_time[0])
+
+p = pd.Period('2017-01-01', 'D')
+pi = pd.PeriodIndex([p])
+
+print(pd.Series(p).dt.end_time[0])
+print(pd.Series(pi).dt.end_time[0])
+print(p.end_time) # x
+print(pi.end_time[0]) 
+print(p.to_timestamp(how='end')) # x
+print(pi.to_timestamp(how='end')[0])
+
+        """
+        Indicator for whether the date is the last day of the month.
+
+        Returns
+        -------
+        Series or array
+            For Series, returns a Series with boolean values. For
+            DatetimeIndex, returns a boolean array.
+
+        See Also
+        --------
+        is_month_start : Indicator for whether the date is the first day
+            of the month.
+
+        Examples
+        --------
+        This method is available on Series with datetime values under
+        the ``.dt`` accessor, and directly on DatetimeIndex.
+
+        >>> dates = pd.Series(pd.date_range("2018-02-27", periods=3))
+        >>> dates
+        0   2018-02-27
+        1   2018-02-28
+        2   2018-03-01
+        dtype: datetime64[ns]
+        >>> dates.dt.is_month_end
+        0    False
+        1    True
+        2    False
+        dtype: bool
+
+        >>> idx = pd.date_range("2018-02-27", periods=3)
+        >>> idx.is_month_end
+        array([False,  True, False], dtype=bool)
+        """)
+
+        """
+        Boolean indicator if the date belongs to a leap year.
+
+        A leap year is a year, which has 366 days (instead of 365) including
+        29th of February as an intercalary day.
+        Leap years are years which are multiples of four with the exception
+        of years divisible by 100 but not by 400.
+
+        Returns
+        -------
+        Series or ndarray
+             Booleans indicating if dates belong to a leap year.
+
+        Examples
+        --------
+        This method is available on Series with datetime values under
+        the ``.dt`` accessor, and directly on DatetimeIndex.
+
+        >>> idx = pd.date_range("2012-01-01", "2015-01-01", freq="Y")
+        >>> idx
+        DatetimeIndex(['2012-12-31', '2013-12-31', '2014-12-31'],
+                      dtype='datetime64[ns]', freq='A-DEC')
+        >>> idx.is_leap_year
+        array([ True, False, False], dtype=bool)
+
+        >>> dates = pd.Series(idx)
+        >>> dates_series
+        0   2012-12-31
+        1   2013-12-31
+        2   2014-12-31
+        dtype: datetime64[ns]
+        >>> dates_series.dt.is_leap_year
+        0     True
+        1    False
+        2    False
+        dtype: bool
+        """)

hacking_asfreq.py

@@ -0,0 +1,12 @@
+import pandas as pd
+import pdb
+
+ts = pd.Series([1.1, 1.2, 1.3, 1.4, 1.5], 
+               index=pd.period_range('1/1/1990', freq='W-MON', periods=5))            
+    
+result = ts.resample('B', convention='end').ffill()
+print(result)
+
+#pdb.set_trace()
+expected = result.to_timestamp('B', how='end').asfreq('B', 'ffill').to_period()
+print(expected)

hacking_offsetting.py

@@ -0,0 +1,24 @@
+import pandas as pd
+
+vec = pd.DatetimeIndex([pd.Timestamp('2000-01-05 00:15:00'),
+ pd.Timestamp('2000-01-31 00:23:00'),
+ pd.Timestamp('2000-01-01 00:00:00'),
+ pd.Timestamp('2000-03-31 00:00:00'),
+ pd.Timestamp('2000-02-29 00:00:00'),
+ pd.Timestamp('2000-12-31 00:00:00'),
+ pd.Timestamp('2000-05-15 00:00:00'),
+ pd.Timestamp('2001-06-15 00:00:00')])
+
+print(isinstance(pd.tseries.offsets.Week, pd.DateOffset))
+
+print(vec + pd.tseries.offsets.Week(weekday=3))
+
+print('wanted')
+print([pd.Timestamp('2000-01-06 00:15:00'),
+ pd.Timestamp('2000-02-03 00:23:00'),
+ pd.Timestamp('2000-01-06 00:00:00'),
+ pd.Timestamp('2000-04-06 00:00:00'),
+ pd.Timestamp('2000-03-02 00:00:00'),
+ pd.Timestamp('2001-01-04 00:00:00'),
+ pd.Timestamp('2000-05-18 00:00:00'),
+ pd.Timestamp('2001-06-21 00:00:00')])

hacking_resample.py

@@ -0,0 +1,37 @@
+import pandas as pd
+import numpy as np
+from pandas.util.testing import assert_frame_equal, assert_series_equal
+
+exp_index = pd.date_range('1/1/2001', end='12/31/2009', freq='A-DEC')
+
+df = pd.DataFrame(index=pd.date_range(
+                                      start='2016-02-11 00:00:00',
+                                      end='2016-02-12 23:00:00',
+                                      freq='H'))
+
+#print(df.resample('BH').count())
+
+start = '1/1/1990'
+end = '12/31/1995'
+freq = 'W-FRI'
+rng = pd.period_range(start, end, freq=freq)
+ts = pd.Series(np.random.randn(len(rng)), index=rng)
+print(ts)
+result = ts.resample('B', convention='end').ffill()
+print(result.index)
+expected = result.to_timestamp('B', how='end')
+print(expected.index)
+# import pdb
+# pdb.set_trace()
+expected = expected.asfreq('B', 'ffill').to_period()
+
+print('wanted')
+print(result.head())
+print(result.tail())
+print('received')
+print(expected.head())
+print(expected.tail())
+
+print(all(result == expected))
+
+assert_series_equal(result, expected)

pandas/core/frame.py

@@ -1287,7 +1287,7 @@ def from_records(cls, data, index=None, exclude=None, columns=None,
 
         return cls(mgr)
 
-    def to_records(self, index=True, convert_datetime64=True):
+    def to_records(self, index=True, convert_datetime64=False):
         """
         Convert DataFrame to a NumPy record array.
 
@@ -1298,7 +1298,9 @@ def to_records(self, index=True, convert_datetime64=True):
         ----------
         index : boolean, default True
             Include index in resulting record array, stored in 'index' field.
-        convert_datetime64 : boolean, default True
+        convert_datetime64 : boolean, default False
+            .. deprecated:: 0.23.0
+
             Whether to convert the index to datetime.datetime if it is a
             DatetimeIndex.
 
@@ -1352,6 +1354,13 @@ def to_records(self, index=True, convert_datetime64=True):
                    ('2018-01-01T09:01:00.000000000', 2, 0.75)],
                   dtype=[('index', '<M8[ns]'), ('A', '<i8'), ('B', '<f8')])
         """
+
+        if convert_datetime64:
+            warnings.warn("The 'convert_datetime64' parameter is "
+                          "deprecated and will be removed in a future "
+                          "version",
+                          FutureWarning, stacklevel=2)
+
         if index:
             if is_datetime64_any_dtype(self.index) and convert_datetime64:
                 ix_vals = [self.index.to_pydatetime()]

pandas/tests/frame/test_convert_to.py

@@ -78,7 +78,10 @@ def test_to_records_dt64(self):
         df = DataFrame([["one", "two", "three"],
                         ["four", "five", "six"]],
                        index=date_range("2012-01-01", "2012-01-02"))
-        assert df.to_records()['index'][0] == df.index[0]
+        with tm.assert_produces_warning(FutureWarning):
+            expected = df.index[0]
+            result = df.to_records(convert_datetime64=True)['index'][0]
+            assert expected == result
 
         rs = df.to_records(convert_datetime64=False)
         assert rs['index'][0] == df.index.values[0]