Series[ExtensionArray].astype(str) failing with "has no attribute ravel"

[jorisvandenbossche]

astype(float) and astype(object) are working on an example Series[DecimalArray], but astype(str) is failing (while on the array itself, it works).

This is because we check for string types, and in that case use self.values (which is still an extension array) and not self.get_values(), and afterwards we call .ravel() on the result:

pandas/pandas/core/internals.py

Lines 649 to 664 in 4efb39f

	if issubclass(dtype.type,
	(compat.text_type, compat.string_types)):
	# use native type formatting for datetime/tz/timedelta
	if self.is_datelike:
	values = self.to_native_types()
	# astype formatting
	else:
	values = self.values
	else:
	values = self.get_values(dtype=dtype)
	# _astype_nansafe works fine with 1-d only
	values = astype_nansafe(values.ravel(), dtype, copy=True)

Example:

In [40]: from pandas.tests.extension.decimal.array import DecimalArray, make_data

In [41]: dec_arr = DecimalArray(make_data()[:3])

In [42]: s = pd.Series(dec_arr)

In [43]: s
Out[43]: 
0    0.84287460911336775648550201367470435798168182...
1    0.41192659105851348044780024792999029159545898...
2    0.46823308426791498959573800675570964813232421875
dtype: decimal

In [44]: s.astype(float)
Out[44]: 
0    0.842875
1    0.411927
2    0.468233
dtype: float64

In [45]: s.astype(object)
Out[45]: 
0    0.84287460911336775648550201367470435798168182...
1    0.41192659105851348044780024792999029159545898...
2    0.46823308426791498959573800675570964813232421875
dtype: object

In [47]: s.astype(str)
---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-47-b17e25b1f3f4> in <module>()
----> 1 s.astype(str)

/home/joris/scipy/pandas/pandas/util/_decorators.py in wrapper(*args, **kwargs)
    175                 else:
    176                     kwargs[new_arg_name] = new_arg_value
--> 177             return func(*args, **kwargs)
    178         return wrapper
    179     return _deprecate_kwarg

/home/joris/scipy/pandas/pandas/core/generic.py in astype(self, dtype, copy, errors, **kwargs)
   5001             # else, only a single dtype is given
   5002             new_data = self._data.astype(dtype=dtype, copy=copy, errors=errors,
-> 5003                                          **kwargs)
   5004             return self._constructor(new_data).__finalize__(self)
   5005 

/home/joris/scipy/pandas/pandas/core/internals.py in astype(self, dtype, **kwargs)
   3670 
   3671     def astype(self, dtype, **kwargs):
-> 3672         return self.apply('astype', dtype=dtype, **kwargs)
   3673 
   3674     def convert(self, **kwargs):

/home/joris/scipy/pandas/pandas/core/internals.py in apply(self, f, axes, filter, do_integrity_check, consolidate, **kwargs)
   3537 
   3538             kwargs['mgr'] = self
-> 3539             applied = getattr(b, f)(**kwargs)
   3540             result_blocks = _extend_blocks(applied, result_blocks)
   3541 

/home/joris/scipy/pandas/pandas/core/internals.py in astype(self, dtype, copy, errors, values, **kwargs)
    573     def astype(self, dtype, copy=False, errors='raise', values=None, **kwargs):
    574         return self._astype(dtype, copy=copy, errors=errors, values=values,
--> 575                             **kwargs)
    576 
    577     def _astype(self, dtype, copy=False, errors='raise', values=None,

/home/joris/scipy/pandas/pandas/core/internals.py in _astype(self, dtype, copy, errors, values, klass, mgr, **kwargs)
    662 
    663                 # _astype_nansafe works fine with 1-d only
--> 664                 values = astype_nansafe(values.ravel(), dtype, copy=True)
    665                 values = values.reshape(self.shape)
    666 

AttributeError: 'DecimalArray' object has no attribute 'ravel'

In [49]: dec_arr.astype(str)
Out[49]: 
array(['0.84287460911336775648550201367470435798168182373046875',
       '0.411926591058513480447800247929990291595458984375',
       '0.46823308426791498959573800675570964813232421875'],
      dtype='<U55')