BUG: hour string automatically converted to datetime in groupby apply

[jorisvandenbossche]

(Encountered with same SO question as #5786 and #5788. With following dataframe:

from StringIO import StringIO
s="""2011.05.16,00:00,1.40893
2011.05.16,01:00,1.40760
2011.05.16,02:00,1.40750
2011.05.16,03:00,1.40649
2011.05.17,02:00,1.40893
2011.05.17,03:00,1.40760
2011.05.17,04:00,1.40750
2011.05.17,05:00,1.40649
2011.05.18,02:00,1.40893
2011.05.18,03:00,1.40760
2011.05.18,04:00,1.40750
2011.05.18,05:00,1.40649"""
df = pd.read_csv(StringIO(s), header=None, names=['date', 'time', 'value'])

where all dates or hours are strings. When doing a groupby on the date and then accessing the value in column time where value is maximal, the hour string is converted to a datetime:

In [21]: df.groupby('date').apply(lambda x: x['time'][x['value'].idxmax()])
Out[21]: 
date
2011.05.16   2013-12-29 00:00:00
2011.05.17   2013-12-29 02:00:00
2011.05.18   2013-12-29 02:00:00
dtype: datetime64[ns]

while in 0.12 this remains a string:

In [63]: df.groupby('date').apply(lambda x: x['time'][x['value'].idxmax()])
Out[63]: 
date
2011.05.16    00:00
2011.05.17    02:00
2011.05.18    02:00
dtype: object

I think in this case this conversion can be highly inappropriate (and you have a chance that just because a string is looking like a date/time component it is converted, while this will not always be your intention).