xarray bump from 2023.10.1 fails tests

[brews]

Failure in dependabot PR #174. I'm not sure if the issue is with dscim or the new xarray 2023.11.0 release.

I could use someone's help triaging this and confirming what the issue is.

There is one failing test. Looks like an issue with "OECD Env-Growth" vs "OECD Env-" when it compares coordinates. The full error message from CI:

=================================== FAILURES ===================================
_______________________ test_concatenate_energy_damages ________________________

tmp_path = PosixPath('/tmp/pytest-of-runner/pytest-0/test_concatenate_energy_damage0')
econvars_fixture = <dscim.menu.simple_storage.EconVars object at 0x7fb21566ceb0>
energy_in_netcdf_fixture = None

    def test_concatenate_energy_damages(
        tmp_path,
        econvars_fixture,
        energy_in_netcdf_fixture,
    ):
        """
        Test that concatenate_energy_damages correctly concatenates separate energy damages by batches across SSP-RCP-GCM-IAMs and saves to separate netcdf file by batches
        """
        concatenate_energy_damages(
            input_path=os.path.join(tmp_path, "energy_in_netcdf"),
            save_path=tmp_path,
            ec_cls=econvars_fixture,
        )
    
        batch = ["batch" + str(i) for i in range(0, 15)]
    
        for b in batch:
            ds_out_expected = xr.Dataset(
                {
                    "histclim_rebased": (
                        ["batch", "rcp", "gcm", "model", "ssp", "region", "year"],
                        np.float32(np.full((1, 2, 2, 2, 2, 2, 2), 2 * 1.273526)),
                    ),
                    "delta_rebased": (
                        ["batch", "rcp", "gcm", "model", "ssp", "region", "year"],
                        np.float32(np.full((1, 2, 2, 2, 2, 2, 2), 2 * 1.273526)),
                    ),
                },
                coords={
                    "batch": (["batch"], [b]),
                    "rcp": (["rcp"], ["rcp45", "rcp85"]),
                    "gcm": (["gcm"], ["ACCESS1-0", "GFDL-CM3"]),
                    "model": (["model"], ["IIASA GDP", "OECD Env-Growth"]),
                    "ssp": (["ssp"], ["SSP2", "SSP3"]),
                    "region": (["region"], ["USA.test_region", "ZWE.test_region"]),
                    "year": (["year"], [2010, 2099]),
                },
            )
    
>           xr.testing.assert_equal(
                ds_out_expected,
                xr.open_dataset(os.path.join(tmp_path, f"rebased_{b}.nc4")),
            )
E           AssertionError: Left and right Dataset objects are not equal
E           
E           Differing coordinates:
E           L * model             (model) <U15 'IIASA GDP' 'OECD Env-Growth'
E           R * model             (model) <U9 'IIASA GDP' 'OECD Env-'

tests/test_input_damages.py:875: AssertionError

This is from this run.

Archived log from the failed CI run is here: logs_689.zip

[JMGilbert]

#229 should address this.

[kemccusker]

Resolved by #229