BUG: boolean frames multiplied by floats have dtypes=object

[erbian]

Code Sample, a copy-pastable example if possible

import pandas as pd
(pd.DataFrame(True, list('ab'), list('cd')) * 1.0).dtypes
#  returns object

type(True * 1.0)
# returns float

Problem description

A boolean frame multiplied by a float should return a DataFrame of floats to be consistent with python scalar operations (e.g., bool * float -> float). Instead it returns dtypes = object.

Prior to 0.21.0, I believe this was the case?

Output of pd.show_versions()

INSTALLED VERSIONS

commit: None
python: 2.7.12.final.0
python-bits: 64
OS: Linux
OS-release: 4.4.0-36-generic
machine: x86_64
processor: x86_64
byteorder: little
LC_ALL: None
LANG: None
LOCALE: None.None

pandas: 0.21.0
pytest: 3.2.5
pip: 9.0.1
setuptools: 37.0.0
Cython: 0.27.3
numpy: 1.13.3
scipy: 1.0.0
pyarrow: None
xarray: 0.10.0
IPython: 5.5.0
sphinx: None
patsy: 0.4.1
dateutil: 2.6.1
pytz: 2017.3
blosc: None
bottleneck: 1.2.1
tables: None
numexpr: 2.6.4
feather: None
matplotlib: 2.1.0
openpyxl: None
xlrd: None
xlwt: 1.3.0
xlsxwriter: None
lxml: None
bs4: None
html5lib: 1.0b10
sqlalchemy: 1.1.15
pymysql: None
psycopg2: 2.7.3.2 (dt dec pq3 ext lo64)
jinja2: 2.8.1
s3fs: None
fastparquet: None
pandas_gbq: 0.2.0
pandas_datareader: None

[jreback]

yeah this looks like it should coerce. would take a PR to fix.

[kgoehner]

This seems to be resolved. At least as of f7d162b when I tried to reproduce it.

>>> (pd.DataFrame(True, list('ab'), list('cd')) * 1.0).dtypes
c    float64
d    float64
dtype: object

Output of pd.show_versions()

INSTALLED VERSIONS

commit : f7d162b
python : 3.7.4.final.0
python-bits : 64
OS : Darwin
OS-release : 18.7.0
machine : x86_64
processor : i386
byteorder : little
LC_ALL : None
LANG : en_US.UTF-8
LOCALE : en_US.UTF-8

pandas : 0.26.0.dev0+555.gf7d162b18.dirty
numpy : 1.17.2
pytz : 2019.3
dateutil : 2.8.0
pip : 19.2.3
setuptools : 41.4.0
Cython : 0.29.13
pytest : 5.2.1
hypothesis : 4.36.2
sphinx : 2.2.0
blosc : None
feather : None
xlsxwriter : 1.2.1
lxml.etree : 4.4.1
html5lib : 1.0.1
pymysql : None
psycopg2 : None
jinja2 : 2.10.3
IPython : 7.8.0
pandas_datareader: None
bs4 : 4.8.0
bottleneck : 1.2.1
fastparquet : 0.3.2
gcsfs : None
lxml.etree : 4.4.1
matplotlib : 3.1.1
numexpr : 2.7.0
odfpy : None
openpyxl : 3.0.0
pandas_gbq : None
pyarrow : 0.15.0
pytables : None
s3fs : 0.3.4
scipy : 1.3.1
sqlalchemy : 1.3.9
tables : 3.5.1
xarray : 0.13.0
xlrd : 1.2.0
xlwt : 1.3.0
xlsxwriter : 1.2.1

[mroeschke]

Would you like to contribute a regression test @Kazz47?

[kgoehner]

Looks like in some platforms the bool is being coerced to a int32 rather than the expected int64. Any ideas where this might be happening or how to reproduce it?

self = <pandas.tests.arithmetic.test_numeric.TestNumericArraylikeArithmeticWithBool object at 0xd5d8258c>
num = 1, all_arithmetic_functions = <built-in function mod>
box_with_array = <class 'pandas.core.frame.DataFrame'>

    @pytest.mark.parametrize("num", [1.0, 1])
    def test_array_like_bool_and_num_op_coerce(
        self, num, all_arithmetic_functions, box_with_array
    ):
        # GH 18549
        op = all_arithmetic_functions
        expected = [op(num, num)]
        expected = tm.box_expected(expected, box_with_array)
        bool_box = tm.box_expected([True], box_with_array)
        try:
>           tm.assert_equal(expected, op(bool_box, num))
E           AssertionError: Attributes are different
E           
E           Attribute "dtype" are different
E           [left]:  int64
E           [right]: int32

[mroeschke]

Guessing there's a routine that's either calling astype(int) (which default to platform int) or forcing astype(np.int64) somewhere. @jbrockmendel would know more

[jbrockmendel]

I'd guess that it is actually the python integer 1 being coerced to int32 by either numpy or numexpr