to_excel incorrectly sets time to midnight for exported datetime column · Issue #9139 · pandas-dev/pandas (original) (raw)

Hello,

I try this:

import requests
from pandas.io.json import json_normalize
import json
import pandas as pd
response = requests.get("http://api.openweathermap.org/data/2.5/history/station?start=1356220800&end=1356307200&type=hour&id=5530")
df = json_normalize(json.loads(response.text)['list'])
df['dt'] = pd.to_datetime(df['dt'], unit='s')
#df = df.set_index('dt')
print(df)
df.to_excel("out.xls")

It displays:

                    dt  humidity.c      ...       pressure.v  temp.c  temp.ma  \
0  2012-12-23 00:00:00           2      ...           1017.0       2   286.15
1  2012-12-23 01:00:00           2      ...           1017.0       2   286.15
2  2012-12-23 02:00:00           2      ...           1017.0       2   285.15
3  2012-12-23 03:00:00           2      ...           1017.0       2   284.15
4  2012-12-23 04:00:00           2      ...           1017.0       2   283.15
...
19 2012-12-23 19:00:00           1      ...           1016.0       1   282.15
20 2012-12-23 20:00:00           1      ...           1015.0       1   281.15
21 2012-12-23 21:00:00           1      ...           1015.0       1   282.15
22 2012-12-23 22:00:00           1      ...           1015.0       1   281.15
23 2012-12-23 23:00:00           1      ...           1005.0       1   286.15
24 2012-12-24 00:00:00           2      ...           1013.5       2   280.15

but if we look at Excel file dt column looks like

dt
2012-12-23 00:00:00
2012-12-23 00:00:00
2012-12-23 00:00:00
2012-12-23 00:00:00
2012-12-23 00:00:00
...
2012-12-23 00:00:00
2012-12-23 00:00:00
2012-12-23 00:00:00
2012-12-23 00:00:00
2012-12-24 00:00:00

to_csv method is ok

In [1]: pd.__version__
Out[1]: '0.15.2'

Kind regards