Keep attributes across operations - "drop_conflicts" option?

[Sevans711]

What is your issue?

The Problem
When I perform simple operations (e.g., +, -, *, /) on DataArray objects, I expected compatible attributes to be maintained, and incompatible attributes to be removed. Even when playing with the available options for keep_attrs, this behavior does not occur. See also #4763 and #2582.

Here is an example showing what happens in each case:

da1 = xr.DataArray(1, attrs=dict(unit_system='SI', a=7))
da2 = xr.DataArray(2, attrs=dict(unit_system='SI', a=8))
with xr.set_options(keep_attrs='default'):
    result = da1 + da2
    result.attrs    # {}
with xr.set_options(keep_attrs=True):
    result = da1 + da2
    result.attrs    # {'unit_system': 'SI', 'a': 7}

Desired behavior:

    result.attrs    # {'unit_system': 'SI'}

That is to say, I want attributes with conflicting values to be discarded, and all other attributes to be maintained.

Suggested Fix
This is similar to the "drop_conflicts" option for merging attrs, from xarray.merge. I suggest that "drop_conflicts" be added as an option for keep_attrs. For example, I would like to be able to do:

with xr.set_options(keep_attrs="drop_conflicts"):
    result = da1 + da2
    result.attrs     # {'unit_system': 'SI'}

If anyone else is able to add this option into the code, that would be amazing!

Or, if there is already a way to accomplish this behavior, I would appreciate any advice on how to do it.

Output of xr.show_versions()

INSTALLED VERSIONS ------------------ commit: None python: 3.9.13 (main, Aug 25 2022, 18:29:29) [Clang 12.0.0 ] python-bits: 64 OS: Darwin OS-release: 21.6.0 machine: x86_64 processor: i386 byteorder: little LC_ALL: None LANG: en_US.UTF-8 LOCALE: ('en_US', 'UTF-8') libhdf5: 1.10.6 libnetcdf: None

xarray: 2023.8.0
pandas: 1.4.4
numpy: 1.21.5
scipy: 1.9.1
netCDF4: None
pydap: None
h5netcdf: None
h5py: 3.7.0
Nio: None
zarr: 2.13.6
cftime: None
nc_time_axis: None
PseudoNetCDF: None
iris: None
bottleneck: 1.3.5
dask: 2022.7.0
distributed: 2022.7.0
matplotlib: 3.5.2
cartopy: None
seaborn: 0.11.2
numbagg: None
fsspec: 2022.7.1
cupy: None
pint: None
sparse: None
flox: None
numpy_groupies: None
setuptools: 63.4.1
pip: 22.2.2
conda: 22.9.0
pytest: 7.1.2
mypy: None
IPython: 7.31.1
sphinx: 5.0.2

[welcome]

Thanks for opening your first issue here at xarray! Be sure to follow the issue template!
If you have an idea for a solution, we would really welcome a Pull Request with proposed changes.
See the Contributing Guide for more.
It may take us a while to respond here, but we really value your contribution. Contributors like you help make xarray better.
Thank you!