Description
What happened?
xr.where(..., keep_attrs=True) overrides coordinate attributes of the result with global attributes from one of its inputs.
What did you expect to happen?
The coordinate attributes to remain unchanged.
Minimal Complete Verifiable Example
import xarray as xr
x_coord = xr.DataArray(
[1, 2, 3],
dims="x",
attrs={"units": "m"}
)
a = xr.DataArray(
[1, 2, 3],
dims="x",
coords={"x": x_coord},
attrs={"units": "K"}
)
res = xr.where(a > 1, a, 0, keep_attrs=True)
assert res.coords["x"].attrs["units"] == "m" # Fails, overridden with "K"MVCE confirmation
- Minimal example — the example is as focused as reasonably possible to demonstrate the underlying issue in xarray.
- Complete example — the example is self-contained, including all data and the text of any traceback.
- Verifiable example — the example copy & pastes into an IPython prompt or Binder notebook, returning the result.
- New issue — a search of GitHub Issues suggests this is not a duplicate.
Relevant log output
No response
Anything else we need to know?
Not sure if this is a duplicate of #2245. If so, feel free to close :)
Environment
INSTALLED VERSIONS
commit: None
python: 3.8.13 | packaged by conda-forge | (default, Mar 25 2022, 06:04:10)
[GCC 10.3.0]
python-bits: 64
OS: Linux
OS-release: 3.10.0-1160.53.1.el7.x86_64
machine: x86_64
processor: x86_64
byteorder: little
LC_ALL: None
LANG: de_DE.UTF-8
LOCALE: ('de_DE', 'UTF-8')
libhdf5: 1.10.6
libnetcdf: 4.7.4
xarray: 2022.6.0
pandas: 1.5.0
numpy: 1.23.3
scipy: 1.9.1
netCDF4: 1.5.6
pydap: None
h5netcdf: None
h5py: 3.3.0
Nio: None
zarr: 2.12.0
cftime: 1.6.2
nc_time_axis: None
PseudoNetCDF: None
rasterio: 1.2.1
cfgrib: None
iris: None
bottleneck: None
dask: 2022.9.1
distributed: 2022.9.1
matplotlib: 3.6.0
cartopy: None
seaborn: None
numbagg: None
fsspec: 2022.8.2
cupy: None
pint: None
sparse: None
flox: None
numpy_groupies: None
setuptools: 65.3.0
pip: 22.2.2
conda: None
pytest: 7.1.3
IPython: None
sphinx: None