issue_comments

So it looks like changing

https://github.com/pydata/xarray/blob/d293f50f9590251ce09543319d1f0dc760466f1b/xarray/backends/api.py#L1280-L1301

to

```python def _validate_datatypes_for_zarr_append(store, dataset): """DataArray.name and Dataset keys must be a string or None"""

def check_dtype(vname):
    store_dtype = store.get_variables()[vname].dtype
    dataset_dtype = dataset[vname].dtype
    if not store_dtype == dataset_dtype:
        raise ValueError(
            f"Mismatched dtypes for variable {vname} between Zarr store on disk "
            f"and dataset to append. Store has dtype {store_dtype} but dataset to "
            f"append has dtype {dataset_dtype}."
        )

for vname in dataset.data_vars:
    check_dtype(vname)

``` could work?

It seems like what we really want to do is verify that the datatype of the appended data matches the data type on disk.

The data type restriction here seems to date back to the original PR adding support for appending. I turned up this comment that seems to summarize the motivation for this check: https://github.com/pydata/xarray/pull/2706#issuecomment-502481584

I think the original issue was that appending a fixed-width string could be a problem if the fixed-width does not match the width of the existing string dtype stored in Zarr.

This obviously doesn't apply in this case, because you are adding an entirely new variable. So I guess the check could be removed in that case.

Thanks for reporting this @kmsampson. My feeling is that it is a bug...which we can hopefully fix pretty easily!

I just ran into this today as well. I am trying to add a dimensionless variable to an existing Zarr store, to help with CF-compliance (if exporting to netCDF), and I ran into this issue. The dtype of my variable is '|S1', and the error message is printed below:

ValueError: Invalid dtype for data variable: <xarray.DataArray 'crs' ()> array(b'', dtype='|S1')

The relevant code is here

https://github.com/pydata/xarray/blob/d293f50f9590251ce09543319d1f0dc760466f1b/xarray/backends/api.py#L1405-L1406

and here

https://github.com/pydata/xarray/blob/d293f50f9590251ce09543319d1f0dc760466f1b/xarray/backends/api.py#L1280-L1298

What I don't understand is why different validation is needed for the append scenario than for the the write scenario. @shoyer worked on this in #5252, so maybe he has some ideas.