Context
In the section Appending to existing Zarr stores, the example is as follows
import xarray as xr
import dask.array
# Write zarr with empty structure
dummies = dask.array.zeros(30, chunks=10)
ds = xr.Dataset({"foo": ("x", dummies)})
path = "path/to/directory.zarr"
ds.to_zarr(path, compute=False)
# Append
ds = xr.Dataset({"foo": ("x", np.arange(30))})
ds.isel(x=slice(0, 10)).to_zarr(path, region={"x": slice(0, 10)})
Question
The example works fine as long as I know the integer slices of the array.
How do I append if I do not know the required region? That is, if I'm given the result of ds.isel(x=slice(0, 10)) without knowing the region slice(0, 10)?
Possible solutions
In theory, I have all the information from the coordinates. For example, for float coordinates, I could do something like
start_index = (ds['x']>=start).argmax().values.item()
end_index = (ds['x']<=end).argmin().values.item()
# region is slice(start_index, end_index)
to determine the isel/zarr indices.
However, this gets fairly involved when dealing with a dataset with many dimensions and coordinates of various types (float, string, datetime). This makes me wonder if there is a more straightforward way.
