WIP

Author: charles-turner-1

ci/environment.yml

@@ -6,6 +6,7 @@ dependencies:
   - cftime
   - codecov
   - dask >=2024.12
+  - esmvalcore >=2.0.0
   - fastprogress >=1.0.0
   - flaky >=3.8.0
   - fsspec >=2024.12

intake_esm/core.py

@@ -5,6 +5,13 @@
 import warnings
 from copy import deepcopy
 
+import esmvalcore.dataset
+
+if typing.TYPE_CHECKING:
+    import esmvalcore
+    from esmvalcore.dataset import Dataset
+    from esmvalcore.typing import FacetValue
+
 import dask
 import packaging.version
 import xarray as xr
@@ -22,6 +29,7 @@
 from fastprogress.fastprogress import progress_bar
 from intake.catalog import Catalog
 
+from ._imports import _ESMVALCORE_AVAILABLE
 from .cat import ESMCatalogModel
 from .derived import DerivedVariableRegistry, default_registry
 from .source import ESMDataSource
@@ -839,6 +847,53 @@ def to_dask(self, **kwargs) -> xr.Dataset:
         _, ds = res.popitem()
         return ds
 
+    def to_iris(
+        self,
+        facets: dict[FacetValue, str],
+        cmorizer: typing.Any | None = None,
+        **kwargs,
+    ) -> esmvalcore.dataset.Dataset:
+        """
+        Convert result to an ESMValCore Dataset.
+
+        This is only possible if the search returned exactly one result.
+
+        Parameters
+        ----------
+        facets: dict[FacetValue, str]
+            Mapping of ESMValCore Dataset facets to their corresponding esm_datastore
+            attributes. For example, the mapping for a dataset containing keys
+            'activity_id', 'source_id', 'member_id', 'experiment_id' would look like:
+            ```python
+             facets =  {
+                       "activity": "activity_id",
+                       "dataset": "source_id",
+                       "ensemble": "member_id",
+                       "exp": "experiment_id",
+                       "grid": "grid_label",
+                        },
+            ```
+        cmorize: Any, optional
+            CMORizer to use in order to CMORize the datastore search results for
+            the ESMValCore Dataset. Presumably this will be a callable? If not set,
+            no CMORization will be done.
+        kwargs: dict
+            TBC.
+        """
+        if not _ESMVALCORE_AVAILABLE:
+            raise ImportError(
+                '`to_iris()` requires the esmvalcore package to be installed. '
+                'To proceed please install esmvalcore using: '
+                ' `python -m pip install esmvalcore` or `conda install -c conda-forge esmvalcore`.'
+            )
+
+        if len(self) != 1:  # quick check to fail more quickly if there are many results
+            raise ValueError(
+                f'Expected exactly one dataset. Received {len(self)} datasets. Please refine your search.'
+            )
+
+        ds = Dataset(**facets)
+
     def _create_derived_variables(self, datasets, skip_on_error):
         if len(self.derivedcat) > 0:
             datasets = self.derivedcat.update_datasets(

tests/test_core.py

@@ -631,3 +631,25 @@ def test__get_threaded(mock_get_env, threaded, ITK_ESM_THREADING, expected):
             intake_esm.core._get_threaded(threaded)
     else:
         assert intake_esm.core._get_threaded(threaded) == expected
+
+
+@mock.patch('intake_esm.core._ESMVALCORE_AVAILABLE', False)
+def test_to_iris_unavailable():
+    cat = intake.open_esm_datastore(zarr_cat_pangeo_cmip6)
+    cat_sub = cat.search(
+        **dict(
+            variable_id=['pr'],
+            experiment_id='ssp370',
+            activity_id='AerChemMIP',
+            source_id='BCC-ESM1',
+            table_id='Amon',
+            grid_label='gn',
+        )
+    )
+    with pytest.raises(ImportError, match=r'`to_iris\(\)` requires the esmvalcore package'):
+        _ = cat_sub.to_iris(
+            xarray_open_kwargs={
+                'consolidated': True,
+                'backend_kwargs': {'storage_options': {'token': 'anon'}},
+            }
+        )