Bug when assigning timedelta64 to a series (or dataframe) with an indexer

[Talmaj]

Code Sample, a copy-pastable example if possible

In [2]: import pandas as pd

In [3]: import numpy as np

In [4]: s = pd.Series(10*[np.timedelta64(10, 'm')])

In [5]: s
Out[5]: 
0   00:10:00
1   00:10:00
2   00:10:00
3   00:10:00
4   00:10:00
5   00:10:00
6   00:10:00
7   00:10:00
8   00:10:00
9   00:10:00
dtype: timedelta64[ns]

In [6]: s.ix[[1,2,3]] = np.timedelta64(20, 'm')

In [7]: s
Out[7]: 
0          00:10:00
1   00:00:00.000000
2   00:00:00.000000
3   00:00:00.000000
4          00:10:00
5          00:10:00
6          00:10:00
7          00:10:00
8          00:10:00
9          00:10:00
dtype: timedelta64[ns]

Expected Output

In [7]: s
Out[7]: 
0          00:10:00
1          00:20:00
2          00:20:00
3          00:20:00
4          00:10:00
5          00:10:00
6          00:10:00
7          00:10:00
8          00:10:00
9          00:10:00
dtype: timedelta64[ns]

output of pd.show_versions()

INSTALLED VERSIONS
------------------
commit: None
python: 3.5.0.final.0
python-bits: 64
OS: Darwin
OS-release: 14.0.0
machine: x86_64
processor: i386
byteorder: little
LC_ALL: en_US.UTF-8
LANG: en_US.UTF-8

pandas: 0.18.1
nose: None
pip: 8.1.2
setuptools: 18.2
Cython: 0.24.1
numpy: 1.11.1
scipy: 0.18.0
statsmodels: None
xarray: None
IPython: 4.0.3
sphinx: 1.4.6
patsy: None
dateutil: 2.5.3
pytz: 2016.6.1
blosc: None
bottleneck: None
tables: 3.2.3.1
numexpr: 2.6.1
matplotlib: 1.5.1
openpyxl: 2.3.3
xlrd: 0.9.4
xlwt: None
xlsxwriter: None
lxml: 3.5.0
bs4: 4.4.1
html5lib: None
httplib2: None
apiclient: None
sqlalchemy: 1.0.15
pymysql: 0.7.5.None
psycopg2: None
jinja2: 2.8
boto: 2.42.0
pandas_datareader: None

[jreback]

the scalar is not getting wrapped in a Timedelta, which handles all of the conversions. you can do this as a work-around. This is also the fix for the bug. PR's are appreciated. here, should be like the section just below. (for timedelta)

In [11]: s.ix[[1,2,3]] = pd.Timedelta(np.timedelta64(20, 'm'))

In [12]: s
Out[12]: 
0   00:10:00
1   00:20:00
2   00:20:00
3   00:20:00
4   00:10:00
5   00:10:00
6   00:10:00
7   00:10:00
8   00:10:00
9   00:10:00
dtype: timedelta64[ns]

[rsmith54]

Hi,

I have submitted a pull request #14160 which fixes this issue. This would be my first contribution to pandas, so if there are any issues let me know. In particular, I looked a bit where to add a test (maybe here ? ), but I wasn't exactly sure how to test this change. Thanks !

Cheers,
Russell