Zarr store array dtype changes for empty object string

[tomwhite]

What happened?

Saving a dataset with an empty object (string) variable to zarr changes the dtype to float64.

What did you expect to happen?

The dtype should remain as O, just as it does for non-empty object variables.

Minimal Complete Verifiable Example

import numpy as np
import xarray as xr
ds = xr.Dataset({"a": np.array([], dtype="O")})
ds["a"].dtype
# prints: dtype('O')
ds.to_zarr("a.zarr")
ds = xr.open_dataset("a.zarr")
ds["a"].dtype
# prints: dtype('float64')

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?

No response

Environment

INSTALLED VERSIONS

commit: None
python: 3.8.13 (default, Mar 28 2022, 06:16:26)
[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_GB.UTF-8
LOCALE: ('en_GB', 'UTF-8')
libhdf5: None
libnetcdf: None

xarray: 2022.11.0
pandas: 1.5.0
numpy: 1.23.4
scipy: 1.9.2
netCDF4: None
pydap: None
h5netcdf: None
h5py: None
Nio: None
zarr: 2.13.3
cftime: None
nc_time_axis: None
PseudoNetCDF: None
rasterio: None
cfgrib: None
iris: None
bottleneck: None
dask: 2022.10.0
distributed: 2022.10.0
matplotlib: 3.6.1
cartopy: None
seaborn: 0.12.0
numbagg: None
fsspec: 2022.8.2
cupy: None
pint: None
sparse: None
flox: None
numpy_groupies: None
setuptools: 63.4.1
pip: 22.2.2
conda: None
pytest: 7.1.3
IPython: 8.5.0
sphinx: 4.2.0

[tomwhite]

This behaviour stems from this part of _infer_dtype where empty object arrays are converted to float arrays:

xarray/xarray/conventions.py

Lines 156 to 157 in 3aa75c8

	if array.size == 0:
	return np.dtype(float)

Is there any reason we couldn't return strings.create_vlen_dtype(str) instead?

[kmuehlbauer]

@tomwhite Sorry for the delay here. I'll respond shortly on your PR #7862, but we might have to reiterate here later