Add Butler.get_dataset

python/lsst/daf/butler/_butler.py

@@ -42,7 +42,7 @@
 from ._butler_repo_index import ButlerRepoIndex
 from ._config import Config, ConfigSubset
 from ._dataset_existence import DatasetExistence
-from ._dataset_ref import DatasetIdGenEnum, DatasetRef
+from ._dataset_ref import DatasetId, DatasetIdGenEnum, DatasetRef
 from ._dataset_type import DatasetType
 from ._deferredDatasetHandle import DeferredDatasetHandle
 from ._file_dataset import FileDataset
@@ -799,6 +799,23 @@ def get_dataset_type(self, name: str) -> DatasetType:
         """
         raise NotImplementedError()
 
+    @abstractmethod
+    def get_dataset(self, id: DatasetId) -> DatasetRef | None:
+        """Retrieve a Dataset entry.
+
+        Parameters
+        ----------
+        id : `DatasetId`
+            The unique identifier for the dataset.
+
+        Returns
+        -------
+        ref : `DatasetRef` or `None`
+            A ref to the Dataset, or `None` if no matching Dataset
+            was found.
+        """
+        raise NotImplementedError()
+
     @abstractmethod
     def find_dataset(
         self,

python/lsst/daf/butler/direct_butler.py

@@ -55,7 +55,7 @@
 from ._butler_config import ButlerConfig
 from ._config import Config
 from ._dataset_existence import DatasetExistence
-from ._dataset_ref import DatasetIdGenEnum, DatasetRef
+from ._dataset_ref import DatasetId, DatasetIdGenEnum, DatasetRef
 from ._dataset_type import DatasetType
 from ._deferredDatasetHandle import DeferredDatasetHandle
 from ._exceptions import ValidationError
@@ -1322,6 +1322,9 @@ def getURI(
     def get_dataset_type(self, name: str) -> DatasetType:
         return self._registry.getDatasetType(name)
 
+    def get_dataset(self, id: DatasetId) -> DatasetRef | None:
+        return self._registry.getDataset(id)
+
     def find_dataset(
         self,
         datasetType: DatasetType | str,

python/lsst/daf/butler/remote_butler/_remote_butler.py

@@ -40,7 +40,7 @@
 from .._butler_config import ButlerConfig
 from .._config import Config
 from .._dataset_existence import DatasetExistence
-from .._dataset_ref import DatasetIdGenEnum, DatasetRef, SerializedDatasetRef
+from .._dataset_ref import DatasetId, DatasetIdGenEnum, DatasetRef, SerializedDatasetRef
 from .._dataset_type import DatasetType, SerializedDatasetType
 from .._deferredDatasetHandle import DeferredDatasetHandle
 from .._file_dataset import FileDataset
@@ -223,6 +223,11 @@ def get_dataset_type(self, name: str) -> DatasetType:
         response.raise_for_status()
         return DatasetType.from_simple(SerializedDatasetType(**response.json()), universe=self.dimensions)
 
+    def get_dataset(self, id: DatasetId) -> DatasetRef | None:
+        path = f"dataset/{id}"
+        response = self._client.get(self._get_url(path))
+        return DatasetRef.from_simple(SerializedDatasetRef(**response.json()), universe=self.dimensions)
+
     def find_dataset(
         self,
         datasetType: DatasetType | str,

python/lsst/daf/butler/remote_butler/server/_server.py

@@ -30,6 +30,7 @@
 __all__ = ("app", "factory_dependency")
 
 import logging
+import uuid
 from functools import cache
 from typing import Any
 
@@ -107,6 +108,25 @@ def get_dataset_type(
     return datasetType.to_simple()
 
 
+@app.get(
+    "/butler/v1/dataset/{id}",
+    summary="Retrieve this dataset definition.",
+    response_model=SerializedDatasetRef | None,
+    response_model_exclude_unset=True,
+    response_model_exclude_defaults=True,
+    response_model_exclude_none=True,
+)
+def get_dataset(id: uuid.UUID, factory: Factory = Depends(factory_dependency)) -> SerializedDatasetRef | None:
+    """Return a single dataset reference."""
+    butler = factory.create_butler()
+    ref = butler.get_dataset(id)
+    if ref is not None:
+        return ref.to_simple()
+    # This could raise a 404 since id is not found. The standard implementation
+    # get_dataset method returns without error so follow that example here.
+    return ref
+
+
 # Not yet supported: TimeSpan is not yet a pydantic model.
 # collections parameter assumes client-side has resolved regexes.
 @app.post(

tests/test_butler.py

@@ -384,7 +384,7 @@ def runPutGetTest(self, storageClass: StorageClass, datasetTypeName: str) -> Dir
                 with self.assertRaises(FileNotFoundError):
                     butler.get(ref)
                 # Registry shouldn't be able to find it by dataset_id anymore.
-                self.assertIsNone(butler.registry.getDataset(ref.id))
+                self.assertIsNone(butler.get_dataset(ref.id))
 
                 # Do explicit registry removal since we know they are
                 # empty

tests/test_server.py

@@ -109,6 +109,9 @@ def test_find_dataset(self):
         self.assertIsInstance(ref, DatasetRef)
         self.assertEqual(ref.id, uuid.UUID("e15ab039-bc8b-4135-87c5-90902a7c0b22"))
 
+        ref2 = self.butler.get_dataset(ref.id)
+        self.assertEqual(ref2, ref)
+
 
 if __name__ == "__main__":
     unittest.main()