BUG: reindex by both index and columns loses DatetimeIndex frequency

[joeb1415]

Given:

from numpy import ones
from pandas import DataFrame, __version__
df = DataFrame(ones((3, 3)), index=[datetime(2012, 1, 1), datetime(2012, 1, 2), datetime(2012, 1, 3)], columns=['a', 'b', 'c'])

time_freq = date_range('2012-01-01', '2012-01-03', freq='d')
some_cols = ['a', 'b']

We get:

__version__

'0.10.1'

df.reindex(index=time_freq).index.freq

<1 Day>

df.reindex(index=time_freq, columns=some_cols).index.freq

None

df.reindex(index=time_freq).reindex(columns=some_cols).index.freq

<1 Day>

[joeb1415]

still present in 0.11.0

[cpcloud]

this is a bug with equality testing for DatetimeIndexes, freq property wasn't being compared, pr coming in the next few minutes

[cpcloud]

unless of course this breaks a bunch of things then more discussion may be needed

[cpcloud]

and of course it breaks 21 tests in the tseries module....@jreback @wesm should DatetimeIndexes with different frequencies compare unequal? it would seem that a lot of code depends on that not being the case...

[wesm]

I don't think so, index.equals(other_index) should be true if the data is the same. we should preserve metadata wherever possible though

[jreback]

closed by #4610