Address comments

Author: mroeschke

asv_bench/benchmarks/timeseries.py

@@ -386,20 +386,35 @@ def time_format_exact(self):
     def time_format_no_exact(self):
         to_datetime(self.s, format='%d%b%y', exact=False)
 
-    def time_cache_with_unique_seconds_and_unit(self):
-        to_datetime(self.unique_numeric_seconds, unit='s')
+    def time_cache_true_with_unique_seconds_and_unit(self):
+        to_datetime(self.unique_numeric_seconds, unit='s', cache=True)
 
-    def time_cache_with_dup_seconds_and_unit(self):
-        to_datetime(self.dup_numeric_seconds, unit='s')
+    def time_cache_false_with_unique_seconds_and_unit(self):
+        to_datetime(self.unique_numeric_seconds, unit='s', cache=False)
 
-    def time_cache_with_dup_string_dates(self):
-        to_datetime(self.dup_string_dates)
+    def time_cache_true_with_dup_seconds_and_unit(self):
+        to_datetime(self.dup_numeric_seconds, unit='s', cache=True)
 
-    def time_cache_with_dup_string_dates_and_format(self):
-        to_datetime(self.dup_string_dates, format='%Y-%m-%d')
+    def time_cache_false_with_dup_seconds_and_unit(self):
+        to_datetime(self.dup_numeric_seconds, unit='s', cache=False)
 
-    def time_cache_with_dup_string_tzoffset_dates(self):
-        to_datetime(self.dup_string_with_tz)
+    def time_cache_true_with_dup_string_dates(self):
+        to_datetime(self.dup_string_dates, cache=True)
+
+    def time_cache_false_with_dup_string_dates(self):
+        to_datetime(self.dup_string_dates, cache=False)
+
+    def time_cache_true_with_dup_string_dates_and_format(self):
+        to_datetime(self.dup_string_dates, format='%Y-%m-%d', cache=True)
+
+    def time_cache_false_with_dup_string_dates_and_format(self):
+        to_datetime(self.dup_string_dates, format='%Y-%m-%d', cache=False)
+
+    def time_cache_true_with_dup_string_tzoffset_dates(self):
+        to_datetime(self.dup_string_with_tz, cache=True)
+
+    def time_cache_false_with_dup_string_tzoffset_dates(self):
+        to_datetime(self.dup_string_with_tz, cache=False)
 
 
 class Offsets(object):

doc/source/whatsnew/v0.22.0.txt

@@ -70,7 +70,7 @@ Performance Improvements
 ~~~~~~~~~~~~~~~~~~~~~~~~
 
 - Indexers on Series or DataFrame no longer create a reference cycle (:issue:`17956`)
-- Added a keyword argument, ``cache``, to :func:`to_datetime` that improved the performance of converting duplicate datetime arguments (:issue:`11665`)
+- Added a keyword argument, ``cache``, to :func:`to_datetime` that improved the performance of converting duplicate datetime arguments. The ``cache`` keyword is set to ``False`` by default. (:issue:`11665`)
 -
 
 .. _whatsnew_0220.docs:

pandas/core/tools/datetimes.py

@@ -37,7 +37,21 @@ def _guess_datetime_format_for_array(arr, **kwargs):
 
 
 def _maybe_cache(arg, format, cache, tz, _convert_listlike):
-    """Create a cache of unique dates from an array of dates"""
+    """
+    Create a cache of unique dates from an array of dates
+
+    Parameters
+    ----------
+    arg : integer, float, string, datetime, list, tuple, 1-d array of dates
+    format : string, strftime to parse time
+    cache: boolean, whether to convert with cache
+    tz: string, timezone of the dates
+    _convert_listlike: function, conversion function to apply on dates
+
+    Returns
+    -------
+    cache_array: Series, cache of converted, unique dates, can be empty
+    """
     from pandas import Series
     cache_array = Series()
     if cache:
@@ -52,7 +66,22 @@ def _maybe_cache(arg, format, cache, tz, _convert_listlike):
 
 
 def _convert_and_box_cache(arg, cache_array, box, errors, tz, name=None):
-    """Convert array of dates with a cache and box the result"""
+    """
+    Convert array of dates with a cache and box the result
+
+    Parameters
+    ----------
+    arg : integer, float, string, datetime, list, tuple, 1-d array of dates
+    cache_array: Series, cache of converted, unique dates
+    box: boolean, True boxes result as an Index-like
+    errors: string, 'ignore' plus box=True will convert result to Index
+    tz: string, timezone of the dates
+    name: string, default None. name for a DatetimeIndex
+
+    Returns
+    -------
+    result: Index-like if box=True else array-like of converted dates
+    """
     from pandas import Series, DatetimeIndex, Index
     result = Series(arg).map(cache_array)
     if box:

pandas/tests/indexes/datetimes/test_tools.py

@@ -372,10 +372,9 @@ def test_to_datetime_tz_psycopg2(self, cache):
                                     dtype='datetime64[ns, UTC]')
         tm.assert_index_equal(result, expected)
 
-    bool_skip = pytest.mark.skipif(True, reason="GH 18111")
-
-    @pytest.mark.parametrize('cache', [pytest.param(True, marks=bool_skip),
-                                       False])
+    @pytest.mark.parametrize('cache',
+        [pytest.param(True, marks=pytest.mark.skipif(True, reason="GH 18111")),
+                      False])
     def test_datetime_bool(self, cache):
         # GH13176
         with pytest.raises(TypeError):