BUG: Incompatible dype warning when assigning boolean series with logical indexer

[m0nzderr]

Pandas version checks

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of pandas.

• I have confirmed this bug exists on the main branch of pandas.

Reproducible Example

import pandas as pd

data1 = pd.Series([True, True, True], dtype=bool)
data2 = pd.Series([False, False, False], dtype=bool)
condition = pd.Series([False, True, False], dtype=bool)

data1[condition] = data2[condition] # > FutureWarning: Setting an item of incompatible dtype...

Issue Description

The assignment data1[condition] = data2[condition] results in warning claiming incompatible types:

FutureWarning: Setting an item of incompatible dtype is deprecated and will raise an error in
a future version of pandas. Value '[False]' has dtype incompatible with bool, please
explicitly cast to a compatible dtype first. data1[condition] = data2[condition]

Which is clearly not true.
This bug is somewhat related to the one reported in #56600, although the use case is different.
Interestingly, the problem disappears when using another dtype, such as int (code below works as expected, no warnings) :

data1 = pd.Series([1 ,2, 3], dtype=int)
data2 = pd.Series([4, 5, 6], dtype=int)
condition = pd.Series([False, True, False], dtype=bool)

data1[condition] = data2[condition] 

Note: Reproduced on pandas==2.2.0 and 3.0.0.dev0+292.g70f367194a

Expected Behavior

There should be no warning.

Installed Versions

INSTALLED VERSIONS ------------------ commit : 70f3671 python : 3.10.12.final.0 python-bits : 64 OS : Linux OS-release : 5.15.133.1-microsoft-standard-WSL2 Version : #1 SMP Thu Oct 5 21:02:42 UTC 2023 machine : x86_64 processor : x86_64 byteorder : little LC_ALL : None LANG : None LOCALE : en_US.UTF-8

pandas : 3.0.0.dev0+292.g70f367194a
numpy : 1.26.4
pytz : 2024.1
dateutil : 2.8.2
setuptools : 59.6.0
pip : 24.0
Cython : None
pytest : None
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : None
lxml.etree : None
html5lib : None
pymysql : None
psycopg2 : None
jinja2 : 3.1.3
IPython : 8.21.0
pandas_datareader : None
adbc-driver-postgresql: None
adbc-driver-sqlite : None
bs4 : 4.12.3
bottleneck : None
dataframe-api-compat : None
fastparquet : None
fsspec : None
gcsfs : None
matplotlib : 3.8.2
numba : None
numexpr : None
odfpy : None
openpyxl : None
pyarrow : None
pyreadstat : None
python-calamine : None
pyxlsb : None
s3fs : None
scipy : 1.12.0
sqlalchemy : None
tables : None
tabulate : None
xarray : None
xlrd : None
zstandard : None
tzdata : 2023.4
qtpy : None
pyqt5 : None

[phofl]

This actually upcasts to object, which means that the warning is kind of correct, but this should definitely work and continue to work.

I suspect that we do a align under the hood that causes this, cc @MarcoGorelli

[m0nzderr]

@phofl I'm not familiar with the implementation and have no idea about the reasons for the upcast, but, intuitively, I see no reason for that to happen. Both series already have the same dtype, so any conversion would be unnecessary. It is also very counter-intuitive to have different behaviors among primitive types (i.e., no upcast in case of ints or floats, but an upcast in case of bools...).

[SpoopyPillow]

I think what's happening is that when we mask the series, we eventually try to set the items at the index array([1]) to array([False]) (which is the second element of data2). However, this array is an object type. When we try to cast this array to a bool, pandas gives a LossySetitemError, which may be the problem.