Can't compare Dataframe to None

[nmichaud]

import pandas
import numpy
df = pandas.DataFrame(numpy.random.randn(8, 3), index=range(8),columns=['A', 'B', 'C'])
None == df

raises: pandas.core.common.PandasError: DataFrame constructor not properly called!

[nmichaud]

It seems that the comparison operators in https://github.com/pydata/pandas/blob/master/pandas/core/frame.py#L564 should be using numpy's comparison operators (so the None comparison is broadcast properly)

[wesm]

I think this ought to raise a TypeError now (I think you must be using an old version of pandas, I didn't have to make any changes for this):

In [3]: df
Out[3]: 
          A         B         C
0  0.545436 -0.658971 -0.537160
1 -0.919375  1.978925 -0.255735
2 -1.641768  0.852692  0.402625
3 -0.377459  0.856477  0.603272
4 -1.931178 -1.177506 -2.266807
5  0.013705  1.124144  0.304718
6 -1.269860 -0.827188  0.749840
7  0.626440  1.225717  3.689088

In [4]: df == None
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
/home/wesm/code/pandas/<ipython-input-4-c0373cc02cc9> in <module>()
----> 1 df == None

/home/wesm/code/pandas/pandas/core/frame.pyc in f(self, other)
    233             return self._combine_series_infer(other, func)
    234         else:
--> 235             return self._combine_const(other, func)
    236 
    237     f.__name__ = name

/home/wesm/code/pandas/pandas/core/frame.pyc in _combine_const(self, other, func)
   2720         if not isinstance(result_values, np.ndarray):
   2721             raise TypeError('Could not compare %s with DataFrame values'
-> 2722                             % repr(other))
   2723 
   2724         return self._constructor(result_values, index=self.index,

TypeError: Could not compare None with DataFrame values