to_string on DataFrame may print column labels in the wrong order

[craustin]

d = {'a': [1,2,3], 'b':[4,5,6]}
from pandas import DataFrame
df = DataFrame(d)

df.to_string()

Returns:
a b
0 1 4
1 2 5
2 3 6

df.to_string(columns=['b','a'])

Returns:
a b
0 4 1
1 5 2
2 6 3

[wesm]

Good point. Currently the passed columns is only used to filter the columns, not actually specify an order. Obviously this is a bug, will be a simple fix (feel free to submit a pull request w/ tests if I don't get to it in the next couple days)

[craustin]

Yeah. Like you said, the bug is either that it reorders the columns - or that it doesn't also reorder the headers.

[wesm]

Fixed this in the above commit