Skip to content

Unable to provide decode_timedelta to open_virtual_dataset #749

Description

@jbusecke

I just refactored some code to use v2 of virtualizarr and with that probably updated my xarray dependencies too.

I am now getting:

[/home/jovyan/mursst-icechunk-updater/.venv/lib/python3.12/site-packages/virtualizarr/xarray.py:341](https://hub.openveda.cloud/user/jbusecke/lab/tree/mursst-icechunk-updater/notebooks/mursst-icechunk-updater/.venv/lib/python3.12/site-packages/virtualizarr/xarray.py#line=340): FutureWarning: In a future version, xarray will not decode timedelta values based on the presence of a timedelta-like units attribute by default. Instead it will rely on the presence of a timedelta64 dtype attribute, which is now xarray's default way of encoding timedelta64 values. To continue decoding timedeltas based on the presence of a timedelta-like units attribute, users will need to explicitly opt-in by passing True or CFTimedeltaCoder(decode_via_units=True) to decode_timedelta. To silence this warning, set decode_timedelta to True, False, or a 'CFTimedeltaCoder' instance.

I tried to provide the argument as suggested by the warning

open_virtual_mfdataset(..., decode_timedelta=True)

but that kwarg is not accepted in the open_virtual_dataset function. Should that be added as an explicit kwarg here? Or should we pass arbitrary kwargs to the xr.open_zarr() call?

Id be happy to work on a PR for this, but wanted to first get a rough idea if I am on the right path.

Happy to provide a MRE if needed, this was just a fairly elaborate workflow, and I wanted to make some more progress.

Metadata

Metadata

Assignees

No one assigned

    Labels

    xarrayRequires changes to xarray upstream

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions