lsst-daf-butler 29.0.1rc1__py3-none-any.whl → 29.1.0rc1__py3-none-any.whl

lsst/daf/butler/__init__.py

@@ -38,6 +38,7 @@ from . import ddl, time_utils
 from ._butler import *
 from ._butler_collections import *
 from ._butler_config import *
+from ._butler_metrics import *
 from ._butler_repo_index import *
 from ._collection_type import CollectionType
 from ._column_categorization import *

lsst/daf/butler/_butler.py

@@ -46,6 +46,7 @@ from lsst.utils.logging import getLogger
 from ._butler_collections import ButlerCollections
 from ._butler_config import ButlerConfig, ButlerType
 from ._butler_instance_options import ButlerInstanceOptions
+from ._butler_metrics import ButlerMetrics
 from ._butler_repo_index import ButlerRepoIndex
 from ._config import Config, ConfigSubset
 from ._exceptions import EmptyQueryResultError, InvalidQueryError
@@ -89,6 +90,10 @@ class SpecificButlerDataset:
     dataset: DatasetRef | None
 
 
+class _DeprecatedDefault:
+    """Default value for a deprecated parameter."""
+
+
 class Butler(LimitedButler):  # numpydoc ignore=PR02
     """Interface for data butler and factory for Butler instances.
 
@@ -154,6 +159,7 @@ class Butler(LimitedButler):  # numpydoc ignore=PR02
         writeable: bool | None = None,
         inferDefaults: bool = True,
         without_datastore: bool = False,
+        metrics: ButlerMetrics | None = None,
         **kwargs: Any,
     ) -> Butler:
         if cls is Butler:
@@ -165,6 +171,7 @@ class Butler(LimitedButler):  # numpydoc ignore=PR02
                 writeable=writeable,
                 inferDefaults=inferDefaults,
                 without_datastore=without_datastore,
+                metrics=metrics,
                 **kwargs,
             )
 
@@ -183,6 +190,7 @@ class Butler(LimitedButler):  # numpydoc ignore=PR02
         writeable: bool | None = None,
         inferDefaults: bool = True,
         without_datastore: bool = False,
+        metrics: ButlerMetrics | None = None,
         **kwargs: Any,
     ) -> Butler:
         """Create butler instance from configuration.
@@ -230,6 +238,8 @@ class Butler(LimitedButler):  # numpydoc ignore=PR02
         without_datastore : `bool`, optional
             If `True` do not attach a datastore to this butler. Any attempts
             to use a datastore will fail.
+        metrics : `ButlerMetrics` or `None`, optional
+            Metrics object to record butler usage statistics.
         **kwargs : `Any`
             Default data ID key-value pairs.  These may only identify
             "governor" dimensions like ``instrument`` and ``skymap``.
@@ -300,6 +310,7 @@ class Butler(LimitedButler):  # numpydoc ignore=PR02
         # passing the "butler" parameter to its constructor.  This has
         # been moved out of the constructor into Butler.clone().
         butler = kwargs.pop("butler", None)
+        metrics = metrics if metrics is not None else ButlerMetrics()
         if butler is not None:
             if not isinstance(butler, Butler):
                 raise TypeError("'butler' parameter must be a Butler instance")
@@ -307,10 +318,17 @@ class Butler(LimitedButler):  # numpydoc ignore=PR02
                 raise TypeError(
                     "Cannot pass 'config', 'searchPaths', or 'writeable' arguments with 'butler' argument."
                 )
-            return butler.clone(collections=collections, run=run, inferDefaults=inferDefaults, dataId=kwargs)
+            return butler.clone(
+                collections=collections, run=run, inferDefaults=inferDefaults, metrics=metrics, dataId=kwargs
+            )
 
         options = ButlerInstanceOptions(
-            collections=collections, run=run, writeable=writeable, inferDefaults=inferDefaults, kwargs=kwargs
+            collections=collections,
+            run=run,
+            writeable=writeable,
+            inferDefaults=inferDefaults,
+            metrics=metrics,
+            kwargs=kwargs,
         )
 
         # Load the Butler configuration.  This may involve searching the
@@ -1274,7 +1292,13 @@ class Butler(LimitedButler):  # numpydoc ignore=PR02
         raise NotImplementedError()
 
     @abstractmethod
-    def removeRuns(self, names: Iterable[str], unstore: bool = True) -> None:
+    def removeRuns(
+        self,
+        names: Iterable[str],
+        unstore: bool | type[_DeprecatedDefault] = _DeprecatedDefault,
+        *,
+        unlink_from_chains: bool = False,
+    ) -> None:
         """Remove one or more `~CollectionType.RUN` collections and the
         datasets within them.
 
@@ -1287,7 +1311,13 @@ class Butler(LimitedButler):  # numpydoc ignore=PR02
             they are present, and attempt to rollback the registry deletions if
             datastore deletions fail (which may not always be possible).  If
             `False`, datastore records for these datasets are still removed,
-            but any artifacts (e.g. files) will not be.
+            but any artifacts (e.g. files) will not be. This parameter is now
+            deprecated and no longer has any effect. Files are always deleted
+            from datastores unless they were ingested using full URIs.
+        unlink_from_chains : `bool`, optional
+            If `True` remove the RUN collection from any chains prior to
+            removing the RUN. If `False` the removal will fail if any chains
+            still refer to the RUN.
 
         Raises
         ------
@@ -1360,7 +1390,14 @@ class Butler(LimitedButler):  # numpydoc ignore=PR02
         raise NotImplementedError()
 
     @abstractmethod
-    def ingest_zip(self, zip_file: ResourcePathExpression, transfer: str = "auto") -> None:
+    def ingest_zip(
+        self,
+        zip_file: ResourcePathExpression,
+        transfer: str = "auto",
+        *,
+        transfer_dimensions: bool = False,
+        dry_run: bool = False,
+    ) -> None:
         """Ingest a Zip file into this butler.
 
         The Zip file must have been created by `retrieve_artifacts_zip`.
@@ -1371,10 +1408,17 @@ class Butler(LimitedButler):  # numpydoc ignore=PR02
             Path to the Zip file.
         transfer : `str`, optional
             Method to use to transfer the Zip into the datastore.
+        transfer_dimensions : `bool`, optional
+            If `True`, dimension record data associated with the new datasets
+            will be transferred from the Zip file, if present.
+        dry_run : `bool`, optional
+            If `True` the ingest will be processed without any modifications
+            made to the target butler and as if the target butler did not
+            have any of the datasets.
 
         Notes
         -----
-        Run collections are created as needed.
+        Run collections and dataset types are created as needed.
         """
         raise NotImplementedError()
 
@@ -1741,8 +1785,8 @@ class Butler(LimitedButler):  # numpydoc ignore=PR02
             warn_limit = True
         with self.query() as query:
             result = (
-                query.where(data_id, where, bind=bind, **kwargs)
-                .data_ids(dimensions)
+                query.data_ids(dimensions)
+                .where(data_id, where, bind=bind, **kwargs)
                 .order_by(*ensure_iterable(order_by))
                 .limit(query_limit)
             )
@@ -1979,8 +2023,8 @@ class Butler(LimitedButler):  # numpydoc ignore=PR02
             warn_limit = True
         with self.query() as query:
             result = (
-                query.where(data_id, where, bind=bind, **kwargs)
-                .dimension_records(element)
+                query.dimension_records(element)
+                .where(data_id, where, bind=bind, **kwargs)
                 .order_by(*ensure_iterable(order_by))
                 .limit(query_limit)
             )
@@ -2123,6 +2167,7 @@ class Butler(LimitedButler):  # numpydoc ignore=PR02
         run: str | None | EllipsisType = ...,
         inferDefaults: bool | EllipsisType = ...,
         dataId: dict[str, str] | EllipsisType = ...,
+        metrics: ButlerMetrics | None = None,
     ) -> Butler:
         """Return a new Butler instance connected to the same repository
         as this one, optionally overriding ``collections``, ``run``,
@@ -2142,5 +2187,7 @@ class Butler(LimitedButler):  # numpydoc ignore=PR02
         dataId : `str`
             Same as ``kwargs`` passed to the constructor.  If omitted, copies
             values from original object.
+        metrics : `ButlerMetrics` or `None`, optional
+            Metrics object to record butler statistics.
         """
         raise NotImplementedError()

lsst/daf/butler/_butler_collections.py

@@ -108,6 +108,10 @@ class ButlerCollections(ABC, Sequence):
         """Collection defaults associated with this butler."""
         raise NotImplementedError("Defaults must be implemented by a subclass")
 
+    def __str__(self) -> str:
+        """Return string representation."""
+        return f"{self.__class__.__name__}(defaults={self.defaults})"
+
     @abstractmethod
     def extend_chain(self, parent_collection_name: str, child_collection_names: str | Iterable[str]) -> None:
         """Add children to the end of a CHAINED collection.

lsst/daf/butler/_butler_instance_options.py

@@ -30,6 +30,8 @@ __all__ = ("ButlerInstanceOptions",)
 import dataclasses
 from typing import Any
 
+from ._butler_metrics import ButlerMetrics
+
 
 @dataclasses.dataclass(frozen=True)
 class ButlerInstanceOptions:
@@ -43,4 +45,5 @@ class ButlerInstanceOptions:
     run: str | None = None
     writeable: bool | None = None
     inferDefaults: bool = True
+    metrics: ButlerMetrics = dataclasses.field(default_factory=ButlerMetrics)
     kwargs: dict[str, Any] = dataclasses.field(default_factory=dict)

lsst/daf/butler/_butler_metrics.py

@@ -0,0 +1,117 @@
+# This file is part of daf_butler.
+#
+# Developed for the LSST Data Management System.
+# This product includes software developed by the LSST Project
+# (http://www.lsst.org).
+# See the COPYRIGHT file at the top-level directory of this distribution
+# for details of code ownership.
+#
+# This software is dual licensed under the GNU General Public License and also
+# under a 3-clause BSD license. Recipients may choose which of these licenses
+# to use; please see the files gpl-3.0.txt and/or bsd_license.txt,
+# respectively.  If you choose the GPL option then the following text applies
+# (but note that there is still no warranty even if you opt for BSD instead):
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+from __future__ import annotations
+
+from collections.abc import Callable, Iterator
+from contextlib import contextmanager
+
+from pydantic import BaseModel
+
+from lsst.utils.logging import LsstLoggers
+from lsst.utils.timer import time_this
+
+
+class ButlerMetrics(BaseModel):
+    """Metrics collected during Butler operations."""
+
+    time_in_put: float = 0.0
+    """Wall-clock time, in seconds, spent in put()."""
+
+    time_in_get: float = 0.0
+    """Wall-clock time, in seconds, spent in get()."""
+
+    n_get: int = 0
+    """Number of datasets retrieved with get()."""
+
+    n_put: int = 0
+    """Number of datasets stored with put()."""
+
+    def reset(self) -> None:
+        """Reset all metrics."""
+        self.time_in_put = 0.0
+        self.time_in_get = 0.0
+        self.n_get = 0
+        self.n_put = 0
+
+    def increment_get(self, duration: float) -> None:
+        """Increment time for get().
+
+        Parameters
+        ----------
+        duration : `float`
+            Duration to add to the get() statistics.
+        """
+        self.time_in_get += duration
+        self.n_get += 1
+
+    def increment_put(self, duration: float) -> None:
+        """Increment time for put().
+
+        Parameters
+        ----------
+        duration : `float`
+            Duration to add to the put() statistics.
+        """
+        self.time_in_put += duration
+        self.n_put += 1
+
+    @contextmanager
+    def _timer(
+        self, handler: Callable[[float], None], log: LsstLoggers | None = None, msg: str | None = None
+    ) -> Iterator[None]:
+        with time_this(log=log, msg=msg) as timer:
+            yield
+        handler(timer.duration)
+
+    @contextmanager
+    def instrument_get(self, log: LsstLoggers | None = None, msg: str | None = None) -> Iterator[None]:
+        """Run code and increment get statistics.
+
+        Parameters
+        ----------
+        log : `logging.Logger` or `None`
+            Logger to use for any timing information.
+        msg : `str` or `None`
+            Any message to be included in log output.
+        """
+        with self._timer(self.increment_get, log=log, msg=msg):
+            yield
+
+    @contextmanager
+    def instrument_put(self, log: LsstLoggers | None = None, msg: str | None = None) -> Iterator[None]:
+        """Run code and increment put statistics.
+
+        Parameters
+        ----------
+        log : `logging.Logger` or `None`
+            Logger to use for any timing information.
+        msg : `str` or `None`
+            Any message to be included in log output.
+        """
+        with self._timer(self.increment_put, log=log, msg=msg):
+            yield

lsst/daf/butler/_config.py

@@ -1254,7 +1254,7 @@ class ConfigSubset(Config):
 
         Global defaults, at lowest priority, are found in the ``config``
         directory of the butler source tree. Additional defaults can be
-        defined using the environment variable ``$DAF_BUTLER_CONFIG_PATHS``
+        defined using the environment variable ``$DAF_BUTLER_CONFIG_PATH``
         which is a PATH-like variable where paths at the front of the list
         have priority over those later.
 

lsst/daf/butler/_dataset_ref.py

@@ -66,8 +66,11 @@ from ._named import NamedKeyDict
 from .datastore.stored_file_info import StoredDatastoreItemInfo
 from .dimensions import (
     DataCoordinate,
+    DimensionDataAttacher,
+    DimensionDataExtractor,
     DimensionGroup,
     DimensionUniverse,
+    SerializableDimensionData,
     SerializedDataCoordinate,
     SerializedDataId,
 )
@@ -907,6 +910,62 @@ class MinimalistSerializableDatasetRef(pydantic.BaseModel):
     data_id: SerializedDataId
     """Data coordinate of this dataset."""
 
+    def to_dataset_ref(
+        self,
+        id: DatasetId,
+        *,
+        dataset_type: DatasetType,
+        universe: DimensionUniverse,
+        attacher: DimensionDataAttacher | None = None,
+    ) -> DatasetRef:
+        """Convert serialized object to a `DatasetRef`.
+
+        Parameters
+        ----------
+        id : `DatasetId`
+            UUID identifying the dataset.
+        dataset_type : `DatasetType`
+            `DatasetType` record corresponding to the dataset type name in the
+            serialized object.
+        universe : `DimensionUniverse`
+            Dimension universe for the dataset.
+        attacher : `DimensionDataAttacher`, optional
+            If provided, will be used to add dimension records to the
+            deserialized `DatasetRef` instance.
+
+        Returns
+        -------
+        ref : `DatasetRef`
+            The deserialized object.
+        """
+        assert dataset_type.name == self.dataset_type_name, (
+            "Given DatasetType does not match the serialized dataset type name"
+        )
+        simple_data_id = SerializedDataCoordinate(dataId=self.data_id)
+        data_id = DataCoordinate.from_simple(simple=simple_data_id, universe=universe)
+        if attacher:
+            data_ids = attacher.attach(dataset_type.dimensions, [data_id])
+            data_id = data_ids[0]
+        return DatasetRef(
+            id=id,
+            run=self.run,
+            datasetType=dataset_type,
+            dataId=data_id,
+        )
+
+    @staticmethod
+    def from_dataset_ref(ref: DatasetRef) -> MinimalistSerializableDatasetRef:
+        """Serialize a ``DatasetRef` to a simplified format.
+
+        Parameters
+        ----------
+        ref : `DatasetRef`
+            `DatasetRef` object to serialize.
+        """
+        return MinimalistSerializableDatasetRef(
+            dataset_type_name=ref.datasetType.name, run=ref.run, data_id=dict(ref.dataId.mapping)
+        )
+
 
 class SerializedDatasetRefContainer(pydantic.BaseModel):
     """Serializable model for a collection of DatasetRef.
@@ -938,6 +997,9 @@ class SerializedDatasetRefContainerV1(SerializedDatasetRefContainer):
     compact_refs: dict[uuid.UUID, MinimalistSerializableDatasetRef]
     """Minimal dataset ref information indexed by UUID."""
 
+    dimension_records: SerializableDimensionData | None = None
+    """Dimension record information"""
+
     def __len__(self) -> int:
         """Return the number of datasets in the container."""
         return len(self.compact_refs)
@@ -957,19 +1019,32 @@ class SerializedDatasetRefContainerV1(SerializedDatasetRefContainer):
         universe: DimensionUniverse | None = None
         dataset_types: dict[str, SerializedDatasetType] = {}
         compact_refs: dict[uuid.UUID, MinimalistSerializableDatasetRef] = {}
+        data_ids: list[DataCoordinate] = []
+        dimensions: list[DimensionGroup] = []
         for ref in refs:
-            simple_ref = ref.to_simple()
-            dataset_type = simple_ref.datasetType
-            assert dataset_type is not None  # For mypy
             if universe is None:
                 universe = ref.datasetType.dimensions.universe
-            if (name := dataset_type.name) not in dataset_types:
-                dataset_types[name] = dataset_type
-            data_id = simple_ref.dataId
-            assert data_id is not None  # For mypy
-            compact_refs[simple_ref.id] = MinimalistSerializableDatasetRef(
-                dataset_type_name=name, run=simple_ref.run, data_id=data_id.dataId
+            if (name := ref.datasetType.name) not in dataset_types:
+                dataset_types[name] = ref.datasetType.to_simple()
+            compact_refs[ref.id] = MinimalistSerializableDatasetRef.from_dataset_ref(ref)
+            if ref.dataId.hasRecords():
+                dimensions.append(ref.datasetType.dimensions)
+                data_ids.append(ref.dataId)
+
+        # Extract dimension record metadata if present.
+        dimension_records = None
+        if data_ids and len(compact_refs) == len(data_ids):
+            dimension_group = DimensionGroup.union(*dimensions, universe=universe)
+
+            # Records were attached to all refs. Store them.
+            extractor = DimensionDataExtractor.from_dimension_group(
+                dimension_group,
+                ignore_cached=False,
+                include_skypix=False,
             )
+            extractor.update(data_ids)
+            dimension_records = SerializableDimensionData.from_record_sets(extractor.records.values())
+
         if universe:
             universe_version = universe.version
             universe_namespace = universe.namespace
@@ -982,6 +1057,7 @@ class SerializedDatasetRefContainerV1(SerializedDatasetRefContainer):
             universe_namespace=universe_namespace,
             dataset_types=dataset_types,
             compact_refs=compact_refs,
+            dimension_records=dimension_records,
         )
 
     def to_refs(self, universe: DimensionUniverse) -> list[DatasetRef]:
@@ -1019,15 +1095,22 @@ class SerializedDatasetRefContainerV1(SerializedDatasetRefContainer):
             name: DatasetType.from_simple(dtype, universe=universe)
             for name, dtype in self.dataset_types.items()
         }
+
+        # Dimension records can be attached if available.
+        # We assume that all dimension information was stored.
+        attacher = None
+        if self.dimension_records:
+            attacher = DimensionDataAttacher(
+                deserializers=self.dimension_records.make_deserializers(universe)
+            )
+
         refs: list[DatasetRef] = []
         for id_, minimal in self.compact_refs.items():
-            simple_data_id = SerializedDataCoordinate(dataId=minimal.data_id)
-            data_id = DataCoordinate.from_simple(simple=simple_data_id, universe=universe)
-            ref = DatasetRef(
-                id=id_,
-                run=minimal.run,
-                datasetType=dataset_types[minimal.dataset_type_name],
-                dataId=data_id,
+            ref = minimal.to_dataset_ref(
+                id_,
+                dataset_type=dataset_types[minimal.dataset_type_name],
+                universe=universe,
+                attacher=attacher,
             )
             refs.append(ref)
         return refs

lsst/daf/butler/_file_dataset.py

@@ -27,15 +27,21 @@
 
 from __future__ import annotations
 
-__all__ = ["FileDataset"]
+__all__ = ("FileDataset", "SerializedFileDataset")
 
+import uuid
+from collections.abc import Callable
 from dataclasses import dataclass
-from typing import Any
+from typing import Any, TypeAlias
+
+import pydantic
 
 from lsst.resources import ResourcePath, ResourcePathExpression
 
-from ._dataset_ref import DatasetRef
+from ._dataset_ref import DatasetRef, MinimalistSerializableDatasetRef
+from ._dataset_type import DatasetType
 from ._formatter import FormatterParameter
+from .dimensions import DimensionUniverse
 
 
 @dataclass
@@ -87,3 +93,72 @@ class FileDataset:
         if not isinstance(other, type(self)):
             return NotImplemented
         return str(self.path) < str(other.path)
+
+    def to_simple(self) -> SerializedFileDataset:
+        """
+        Convert this instance to a simplified, JSON-serializable object.
+
+        Returns
+        -------
+        serialized : `SerializedFileDataset`
+            Serializable representation of this `FileDataset` instance.
+        """
+        if self.formatter is None:
+            formatter = None
+        elif isinstance(self.formatter, str):
+            formatter = self.formatter
+        else:
+            formatter = self.formatter.name()
+
+        refs = {ref.id: MinimalistSerializableDatasetRef.from_dataset_ref(ref) for ref in self.refs}
+
+        return SerializedFileDataset(
+            refs=refs,
+            path=str(self.path),
+            formatter=formatter,
+        )
+
+    @staticmethod
+    def from_simple(
+        dataset: SerializedFileDataset, *, dataset_type_loader: DatasetTypeLoader, universe: DimensionUniverse
+    ) -> FileDataset:
+        """
+        Deserialize a `SerializedFileDataset` into a `FileDataset`.
+
+        Parameters
+        ----------
+        dataset : `SerializedFileDataset`
+            Object to deserialize.
+        dataset_type_loader : `Callable` [[ `str` ], `DatasetType` ]
+            Function that takes a string dataset type name as its
+            only parameter, and returns an instance of `DatasetType`.
+            Used to deserialize the `DatasetRef` instances contained
+            in the serialized `FileDataset`.
+        universe : `DimensionUniverse`
+            Dimension universe associated with the `Butler` instance that
+            created the serialized `FileDataset` instance.
+
+        Returns
+        -------
+        file_dataset : `FileDataset`
+            Deserialized equivalent of the input dataset.
+        """
+        refs = [
+            ref.to_dataset_ref(id, universe=universe, dataset_type=dataset_type_loader(ref.dataset_type_name))
+            for id, ref in dataset.refs.items()
+        ]
+        return FileDataset(path=dataset.path, refs=refs, formatter=dataset.formatter)
+
+
+DatasetTypeLoader: TypeAlias = Callable[[str], DatasetType]
+"""Type signature for a function that takes a string dataset type name as its
+only parameter, and returns an instance of `DatasetType`.
+"""
+
+
+class SerializedFileDataset(pydantic.BaseModel):
+    """Serializable format of `FileDataset` object."""
+
+    refs: dict[uuid.UUID, MinimalistSerializableDatasetRef]
+    path: str
+    formatter: str | None = None

lsst/daf/butler/_limited_butler.py

@@ -31,11 +31,13 @@ __all__ = ("LimitedButler",)
 
 import logging
 from abc import ABC, abstractmethod
-from collections.abc import Iterable
+from collections.abc import Iterable, Iterator
+from contextlib import contextmanager
 from typing import Any, ClassVar
 
 from lsst.resources import ResourcePath
 
+from ._butler_metrics import ButlerMetrics
 from ._dataset_provenance import DatasetProvenance
 from ._dataset_ref import DatasetRef
 from ._deferredDatasetHandle import DeferredDatasetHandle
@@ -138,7 +140,8 @@ class LimitedButler(ABC):
         to use a resolved `DatasetRef`. Subclasses can support more options.
         """
         log.debug("Butler get: %s, parameters=%s, storageClass: %s", ref, parameters, storageClass)
-        return self._datastore.get(ref, parameters=parameters, storageClass=storageClass)
+        with self._metrics.instrument_get(log, msg="Retrieved dataset"):
+            return self._datastore.get(ref, parameters=parameters, storageClass=storageClass)
 
     def getDeferred(
         self,
@@ -415,6 +418,30 @@ class LimitedButler(ABC):
         """
         raise NotImplementedError()
 
+    @contextmanager
+    def record_metrics(self, metrics: ButlerMetrics | None = None) -> Iterator[ButlerMetrics]:
+        """Enable new metrics recording context.
+
+        Parameters
+        ----------
+        metrics : `lsst.daf.butler.ButlerMetrics`
+            Optional override metrics object. If given, this will be the
+            same object returned by the context manager.
+
+        Yields
+        ------
+        metrics : `lsst.daf.butler.ButlerMetrics`
+            Metrics recorded within this context. This temporarily replaces
+            any existing metrics object associated with this butler.
+        """
+        old_metrics = self._metrics
+        new_metrics = metrics if metrics is not None else ButlerMetrics()
+        try:
+            self._metrics = new_metrics
+            yield new_metrics
+        finally:
+            self._metrics = old_metrics
+
     @property
     @abstractmethod
     def dimensions(self) -> DimensionUniverse:
@@ -430,3 +457,8 @@ class LimitedButler(ABC):
     """An object that maps known storage class names to objects that fully
     describe them (`StorageClassFactory`).
     """
+
+    _metrics: ButlerMetrics
+    """An object for recording metrics associated with this butler.
+    (`ButlerMetrics`)
+    """