Fix #12037 Error when Resampling using pd.tseries.offsets.Nano as period

Closes #12037 Author: Bran Yang <snowolfy@163.com> Closes #12270 from BranYang/nanosec and squashes the following commits: bff0c85 [Bran Yang] Add to whatsnew and some comments fd0b307 [Bran Yang] Fix #12037 Error when Resampling using pd.tseries.offsets.Nano as period
pandas-dev · Feb 10, 2016 · ab29f93 · ab29f93
1 parent e9558d3
commit ab29f93
Show file tree

Hide file tree

Showing 3 changed files with 32 additions and 3 deletions.
diff --git a/doc/source/whatsnew/v0.18.0.txt b/doc/source/whatsnew/v0.18.0.txt
@@ -828,6 +828,7 @@ Bug Fixes
 - Bug in ``Series.str.get_dummies`` when one of the variables was 'name' (:issue:`12180`)
 - Bug in ``pd.concat`` while concatenating tz-aware NaT series. (:issue:`11693`, :issue:`11755`)
 - Bug in ``pd.read_stata`` with version <= 108 files (:issue:`12232`)
+- Bug in ``Series.resample`` using a frequency of ``Nano`` when the index is a ``DatetimeIndex`` and contains non-zero nanosecond parts (:issue:`12037`)
 
 
 - Bug in ``Timedelta.round`` with negative values (:issue:`11690`)
@@ -845,7 +846,7 @@ Bug Fixes
 - Bug in ``DataFrame.query`` containing an assignment (:issue:`8664`)
 
 - Bug in ``from_msgpack`` where ``__contains__()`` fails for columns of the unpacked ``DataFrame``, if the ``DataFrame`` has object columns. (:issue:`11880`)
-- Bug in ``df.resample()`` on categorical data with ``TimedeltaIndex`` (:issue:`12169`)
+- Bug in ``.resample`` on categorical data with ``TimedeltaIndex`` (:issue:`12169`)
 
 
 - Bug in timezone info lost when broadcasting scalar datetime to ``DataFrame`` (:issue:`11682`)

diff --git a/pandas/tseries/resample.py b/pandas/tseries/resample.py
@@ -854,9 +854,14 @@ def _get_time_bins(self, ax):
                                        closed=self.closed,
                                        base=self.base)
         tz = ax.tz
+        # GH #12037
+        # use first/last directly instead of call replace() on them
+        # because replace() will swallow the nanosecond part
+        # thus last bin maybe slightly before the end if the end contains
+        # nanosecond part and lead to `Values falls after last bin` error
         binner = labels = DatetimeIndex(freq=self.freq,
-                                        start=first.replace(tzinfo=None),
-                                        end=last.replace(tzinfo=None),
+                                        start=first,
+                                        end=last,
                                         tz=tz,
                                         name=ax.name)
 

diff --git a/pandas/tseries/tests/test_resample.py b/pandas/tseries/tests/test_resample.py
@@ -1236,6 +1236,29 @@ def test_monthly_resample_error(self):
         # it works!
         ts.resample('M')
 
+    def test_nanosecond_resample_error(self):
+        # GH 12307 - Values falls after last bin when
+        # Resampling using pd.tseries.offsets.Nano as period
+        start = 1443707890427
+        exp_start = 1443707890400
+        indx = pd.date_range(
+            start=pd.to_datetime(start),
+            periods=10,
+            freq='100n'
+        )
+        ts = pd.Series(range(len(indx)), index=indx)
+        r = ts.resample(pd.tseries.offsets.Nano(100))
+        result = r.agg('mean')
+
+        exp_indx = pd.date_range(
+            start=pd.to_datetime(exp_start),
+            periods=10,
+            freq='100n'
+        )
+        exp = pd.Series(range(len(exp_indx)), index=exp_indx)
+
+        assert_series_equal(result, exp)
+
     def test_resample_anchored_intraday(self):
         # #1471, #1458