DataFrame.dropna() not working with sparse columns

[rjboczar]

import numpy as np
import pandas as pd

A = pd.DataFrame({'a': [0,1], 'b': pd.SparseArray([np.nan, 1])})

# Prints empty DataFrame
A.dropna()

Not sure if I'm using this correctly, but I'd expect only the first row to be dropped.

Output of pd.show_versions()

INSTALLED VERSIONS

commit : None
python : 3.6.9.final.0
python-bits : 64
OS : Darwin
OS-release : 18.7.0
machine : x86_64
processor : i386
byteorder : little
LC_ALL : None
LANG : en_US.UTF-8
LOCALE : en_US.UTF-8

pandas : 0.25.1
numpy : 1.16.4
pytz : 2019.2
dateutil : 2.8.0
pip : 19.2.2
setuptools : 41.0.1
Cython : None
pytest : None
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : None
lxml.etree : None
html5lib : None
pymysql : None
psycopg2 : None
jinja2 : 2.10.1
IPython : 7.7.0
pandas_datareader: None
bs4 : None
bottleneck : None
fastparquet : None
gcsfs : None
lxml.etree : None
matplotlib : 3.1.1
numexpr : None
odfpy : None
openpyxl : None
pandas_gbq : None
pyarrow : None
pytables : None
s3fs : None
scipy : 1.3.1
sqlalchemy : None
tables : None
xarray : None
xlrd : None
xlwt : None
xlsxwriter : None

[TomAugspurger]

Huh, the bug is in actually in DataFrame.__invert__

In [39]: pd.isna(A)
Out[39]:
       a      b
0  False   True
1  False  False

In [40]: ~pd.isna(A)
Out[40]:
    a   b
0  -1  -2
1  -1  -1

Right now we call com.values_from_object. Ideally we would do some kind of blockwise __invert__ and wrap.

cc @jbrockmendel.

[mroeschke]

This looks to work on master. Could use a test I suppose

In [19]: import numpy as np
    ...: import pandas as pd
    ...:
    ...: A = pd.DataFrame({'a': [0,1], 'b': pd.arrays.SparseArray([np.nan, 1])})
    ...:
    ...: A.dropna()
Out[19]:
   a    b
1  1  1.0

In [23]: pd.__version__
Out[23]: '1.1.0.dev0+1361.g77a0f19c5'

[Dhron]

Hey, I'd like to work on writing a test for this, is that okay? =)

[MBrouns]

take