setting a dataframe row with .ix with columns of different datatypes

[Tiago Requeijo]

Hi,

It seems that doing 

df.ix[label] = [... values ...]

only works when the columns are of similar datatypes. For example

import pandas

data = pandas.DataFrame.from_dict({1: [1,2,3], 2: [3,4,5]})

print data.dtypes

data.ix[1] = [10, 20] # works

print data

print

data = pandas.DataFrame.from_dict({1: [1.,2.,3.], 2: [3.,4.,5.]})

print data.dtypes

data.ix[1] = [10., 20.] # works

print data

print

data = pandas.DataFrame.from_dict({1: [1.,2.,3.], 2: [3,4,5]})

print data.dtypes

data.ix[1] = [10., 20] # fails

print data

print

Is this intended?

Thanks,

Tiago

[Wouter Overmeire]

2012/6/5 Tiago Requeijo <tiago.r...@gmail.com>

I opened an issue for this https://github.com/pydata/pandas/issues/1432.

[Wes McKinney]

This has been fixed