List indexer on PeriodIndex doesn't coerce strings · Issue #11278 · pandas-dev/pandas (original) (raw)

In [2]: index = pd.period_range(start='2000', periods=20, freq='B') In [5]: series = pd.Series(range(20), index=index)

In [6]: series Out[6]: 2000-01-03 0 2000-01-04 1 2000-01-05 2 ... 2000-01-26 17 2000-01-27 18 2000-01-28 19 Freq: B, dtype: int64

In [7]: series.loc['2000-01-14'] Out[7]: 9

Supplying a list of Periods as the indexer works as expected:

In [15]: series[[pd.Period(d, freq='B') for d in ['2000-01-14', '2000-01-18']]] Out[15]: 2000-01-14 9 2000-01-18 11 Freq: B, dtype: int64

But not with strings:

In [10]: series.loc[['2000-01-14', '2000-01-15']]

KeyError Traceback (most recent call last) in () ----> 1 series.loc[['2000-01-14', '2000-01-15']]

/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/pandas/core/indexing.py in getitem(self, key) 1187 return self._getitem_tuple(key) 1188 else: -> 1189 return self._getitem_axis(key, axis=0) 1190 1191 def _getitem_axis(self, key, axis=0):

/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/pandas/core/indexing.py in _getitem_axis(self, key, axis) 1321 raise ValueError('Cannot index with multidimensional key') 1322 -> 1323 return self._getitem_iterable(key, axis=axis) 1324 1325 # nested tuple slicing

/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/pandas/core/indexing.py in _getitem_iterable(self, key, axis) 931 def _getitem_iterable(self, key, axis=0): 932 if self._should_validate_iterable(axis): --> 933 self._has_valid_type(key, axis) 934 935 labels = self.obj._get_axis(axis)

/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/pandas/core/indexing.py in _has_valid_type(self, key, axis) 1269 1270 raise KeyError("None of [%s] are in the [%s]" % -> 1271 (key, self.obj._get_axis_name(axis))) 1272 1273 return True

KeyError: "None of [['2000-01-14', '2000-01-15']] are in the [index]"

It also gives NaN without the loc indexer:

In [16]: series[['2000-01-14', '2000-01-18']] Out[16]: 2000-01-14 NaN 2000-01-18 NaN dtype: float64

When supplied with the ints behind the PeriodIndex, it doesn't resolve a single int, but will resolve a list:

In [4]: index.values Out[4]: array([7827, 7828, 7829, 7830, 7831, 7832, 7833, 7834, 7835, 7836, 7837, 7838, 7839, 7840, 7841, 7842, 7843, 7844, 7845, 7846])

In [9]: series.loc[[7829, 7830]] Out[9]: 7829 2 7830 3 dtype: int64

In [8]: series.loc[7829]

TypeError Traceback (most recent call last) in () ----> 1 series.loc[7829]

/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/pandas/core/indexing.py in getitem(self, key) 1187 return self._getitem_tuple(key) 1188 else: -> 1189 return self._getitem_axis(key, axis=0) 1190 1191 def _getitem_axis(self, key, axis=0):

/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/pandas/core/indexing.py in _getitem_axis(self, key, axis) 1331 1332 # fall thru to straight lookup -> 1333 self._has_valid_type(key, axis) 1334 return self._get_label(key, axis=axis) 1335

/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/pandas/core/indexing.py in _has_valid_type(self, key, axis) 1283 1284 try: -> 1285 key = self._convert_scalar_indexer(key, axis) 1286 if not key in ax: 1287 error()

/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/pandas/core/indexing.py in _convert_scalar_indexer(self, key, axis) 161 ax = self.obj._get_axis(min(axis, self.ndim - 1)) 162 # a scalar --> 163 return ax._convert_scalar_indexer(key, kind=self.name) 164 165 def _convert_slice_indexer(self, key, axis):

/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/pandas/tseries/base.py in _convert_scalar_indexer(self, key, kind) 333 334 if kind in ['loc'] and lib.isscalar(key) and (is_integer(key) or is_float(key)): --> 335 self._invalid_indexer('index',key) 336 337 return super(DatetimeIndexOpsMixin, self)._convert_scalar_indexer(key, kind=kind)

/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/pandas/core/index.py in _invalid_indexer(self, form, key) 942 klass=type(self), 943 key=key, --> 944 kind=type(key))) 945 946 def get_duplicates(self):