datetimeindex returns wrong datetimes with tolist() and values()

[saroele]

In [82]: dr = pandas.date_range(start='2012-01-01', periods=10, freq='30Min')

In [83]: dr
Out[83]:
<class 'pandas.tseries.index.DatetimeIndex'>
[2012-01-01 00:00:00, ..., 2012-01-01 04:30:00]
Length: 10, Freq: 30T, Timezone: None

In [84]: dr.values
Out[84]:
array([1970-01-16 224:00:00, 1970-01-16 224:30:00, 1970-01-16 225:00:00,
1970-01-16 225:30:00, 1970-01-16 226:00:00, 1970-01-16 226:30:00,
1970-01-16 227:00:00, 1970-01-16 227:30:00, 1970-01-16 228:00:00,
1970-01-16 228:30:00], dtype=datetime64[ns])

[wesm]

I'll fix tolist but .values can't be fixed-- it's a bug with how NumPy 1.6 displays nanosecond timestamps. See:

http://pandas.pydata.org/pandas-docs/dev/whatsnew.html#potential-porting-issues-for-pandas-0-7-3-users

[wesm]

Implemented DatetimeIndex.tolist. values will have to stay as is