BUG: construction of tz-naive and tz-aware should not coerce

[jreback]

In [1]: Series([Timestamp('20130101'), Timestamp('20130101', tz='US/Eastern')])
Out[1]: 
0   2012-12-31 19:00:00-05:00
1   2013-01-01 00:00:00-05:00
dtype: datetime64[ns, US/Eastern]

should instead be

In [2]: Series([Timestamp('20130101'), Timestamp('20130101', tz='US/Eastern')], dtype=object)
Out[2]: 
0          2013-01-01 00:00:00
1    2013-01-01 00:00:00-05:00
dtype: object

[theodoreschen]

Would making tz default to 'UTC' upon Timestamp() declaration be sufficient to fix this bug?

In [10]: pd.Series([pd.Timestamp('20130101'), pd.Timestamp('20130101', tz='US/Eastern')], dtype=object)
Out[10]:
0          2013-01-01 00:00:00
1    2013-01-01 00:00:00-05:00
dtype: object

In [11]: pd.Series([pd.Timestamp('20130101', tz='UTC'), pd.Timestamp('20130101', tz='US/Eastern')])
Out[11]:
0    2013-01-01 00:00:00+00:00
1    2013-01-01 00:00:00-05:00
dtype: object

[jreback]

@theairportexplorer no!

naive and tz-aware UTC are different! The default is naive (has always been and always will).

[jamestran201]

The rough code path for this call Series([Timestamp('20130101'), Timestamp('20130101', tz='US/Eastern')]) is:
series.py.Series.__init__ (263): _sanitize_array()
series.py.Series._sanitize_array (3179): maybe_cast_to_datetime()
cast.py.maybe_cast_to_datetime (1044): maybe_infer_to_datetimelike()
cast.py.maybe_infer_to_datetimelike (914): try_datetime
cast.py.try_datetime (882): tslib.array_to_datetime()
tslib.pyx array_to_datetime(): raise ValueError because it's trying convert Timestamp('20130101', tz='US/Eastern') to datetime64 with utc=None?
cast.py.try_datetime(890): tools/datetimes.py.to_datetime()
tools/datetimes.py.to_datetime (466): _convert_listlike
tools/datetimes.py._convert_listlike (361): tslib.array_to_datetime -> ValueError
tools/datetimes.py._conver_listlike (376): tslib.datetime_to_datetime64(arg)
-> array(['2013-01-01T00:00:00.000000000', '2013-01-01T05:00:00:000000000'],dtype=datetime64[ns])