left join on column doesn't work with missing values on right side

[rafaljozefowicz]

df1 = pd.DataFrame({"i1" : [0, 1], "i2" : [0, 1]})
df2 = pd.DataFrame({"i1" : [0], "i3" : [0]})
df1.join(df2, on="i1", rsuffix="_")

throws:

pandas-0.10.0-py2.7-linux-x86_64.egg/pandas/tools/merge.pyc in _maybe_add_join_keys(self, result, left_indexer, right_indexer)
    224                         continue
    225 
--> 226                     left_na_indexer = left_indexer.take(na_indexer)
    227                     key_col.put(na_indexer, com.take_1d(self.left_join_keys[i],
    228                                                         left_na_indexer))

AttributeError: 'NoneType' object has no attribute 'take'

while the following works as expected:

df1 = pd.DataFrame({"i1" : [0, 1], "i2" : [0.5, 1.5]}).set_index("i1")
df2 = pd.DataFrame({"i1" : [0], "i3" : [0.7]}).set_index("i1")
df1.join(df2, rsuffix="_")

[hayd]

I think I have a fix for this, see upcoming pull request.