Loading then saving a VCF imported dataset fails with TypeError: expected unicode string, found 130

[benjeffery]

vcf_to_zarr("example.vcf.gz", "vcf.zarr")
ds = sg.load_dataset("vcf.zarr")
sg.save_dataset(ds, "vcf2.zarr")

Currently fails with
TypeError: expected unicode string, found 130
I believe this is due to pydata/xarray#3476 as the workaround suggested there:

    for v in list(ds.coords.keys()):
        if ds.coords[v].dtype == object:
            ds[v].encoding.clear()

    for v in list(ds.variables.keys()):
        if ds[v].dtype == object:
            ds[v].encoding.clear()

prevents this error. I'm not sure there is anything sgkit can do about this - looking into it.

[tomwhite]

See also #852

[benjeffery]

Ahh, nice! Didn't see that, thanks @tomwhite.

[benjeffery]

I can confirm that using sgkit HEAD fixes this issue.