Dataframe column filled with .sum() of int-series has dtype not int64, but object

[Rik-de-Kort]

MnWE

import pandas as pd

the_int_column = pd.Series([1, 2, 3])
foobar = pd.DataFrame(columns=("ints",))
foobar.loc["First"] = [the_int_column.sum()]

print(foobar.dtypes)

the_float_column = pd.Series([1.0, 2.0, 3.0])
floatbar = pd.DataFrame(columns=("floats",))
floatbar.loc["First"] = [the_float_column.sum()]

print(floatbar.dtypes)

This yields

ints    object
dtype: object
floats    float64
dtype: object

Problem description

Assigning the sum of a Series of (vanilla Python) ints to a DataFrame leads to not specific enough type inference on the frame: it keeps the object type. Vanilla floats work fine, as the second example demonstrates.

This is an issue for some testing I'm doing, where I want to ensure a column of these sums contains only numeric data, and specifically integers.

Might be related to #20635.

Expected Output

Expected output:

ints       int64
dtype: object
floats    float64
dtype: object

Output of pd.show_versions()

INSTALLED VERSIONS ------------------ commit: None python: 3.6.4.final.0 python-bits: 64 OS: Windows OS-release: 10 machine: AMD64 processor: Intel64 Family 6 Model 142 Stepping 10, GenuineIntel byteorder: little LC_ALL: None LANG: None LOCALE: None.None pandas: 0.22.0 pytest: 3.3.2 pip: 9.0.1 setuptools: 38.4.0 Cython: 0.27.3 numpy: 1.14.0 scipy: 1.0.0 pyarrow: None xarray: None IPython: 6.2.1 sphinx: 1.6.6 patsy: 0.5.0 dateutil: 2.6.1 pytz: 2017.3 blosc: None bottleneck: 1.2.1 tables: 3.4.2 numexpr: 2.6.4 feather: None matplotlib: 2.1.2 openpyxl: 2.4.10 xlrd: 1.1.0 xlwt: 1.3.0 xlsxwriter: 1.0.2 lxml: 4.1.1 bs4: 4.6.0 html5lib: 1.0.1 sqlalchemy: 1.2.1 pymysql: None psycopg2: None jinja2: 2.10 s3fs: None fastparquet: None pandas_gbq: None pandas_datareader: None

[jreback]

duplicate issue of #20635