Coverage for python/lsst/daf/butler/datastores/inMemoryDatastore.py: 91%
217 statements
« prev ^ index » next coverage.py v7.17.2, created at 2026-11-06 09:46 +0000
1# This file is part of daf_butler.
2#
3# Developed for the LSST Data Management System.
4# This product includes software developed by the LSST Project
5# (http://www.lsst.org).
6# See the COPYRIGHT file at the top-level directory of this distribution
7# for details of code ownership.
8#
9# This software is dual licensed under the GNU General Public License and also
10# under a 3-clause BSD license. Recipients may choose which of these licenses
11# to use; please see the files gpl-3.0.txt and/or bsd_license.txt,
12# respectively. If you choose the GPL option then the following text applies
13# (but note that there is still no warranty even if you opt for BSD instead):
14#
15# This program is free software: you can redistribute it and/or modify
16# it under the terms of the GNU General Public License as published by
17# the Free Software Foundation, either version 3 of the License, or
18# (at your option) any later version.
19#
20# This program is distributed in the hope that it will be useful,
21# but WITHOUT ANY WARRANTY; without even the implied warranty of
22# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
23# GNU General Public License for more details.
24#
25# You should have received a copy of the GNU General Public License
26# along with this program. If not, see <http://www.gnu.org/licenses/>.
28"""In-memory datastore."""
30from __future__ import annotations
32__all__ = ("InMemoryDatastore", "StoredMemoryItemInfo")
34import logging
35import time
36from collections.abc import Collection, Iterable, Mapping
37from dataclasses import dataclass
38from typing import TYPE_CHECKING, Any
39from urllib.parse import urlencode
41from lsst.daf.butler import DatasetId, DatasetRef, StorageClass
42from lsst.daf.butler._exceptions import DatasetTypeNotSupportedError
43from lsst.daf.butler.datastore import DatasetRefURIs, DatastoreConfig, DuplicateIdError
44from lsst.daf.butler.datastore.generic_base import GenericBaseDatastore, post_process_get
45from lsst.daf.butler.datastore.record_data import DatastoreRecordData
46from lsst.daf.butler.datastore.stored_file_info import StoredDatastoreItemInfo
47from lsst.daf.butler.utils import transactional
48from lsst.resources import ResourcePath, ResourcePathExpression
50if TYPE_CHECKING:
51 from lsst.daf.butler import Config, DatasetProvenance, DatasetType, LookupKey
52 from lsst.daf.butler.datastore import DatastoreOpaqueTable
53 from lsst.daf.butler.datastores.file_datastore.retrieve_artifacts import ArtifactIndexInfo
54 from lsst.daf.butler.registry.interfaces import DatasetIdRef, DatastoreRegistryBridgeManager
56log = logging.getLogger(__name__)
59@dataclass(frozen=True, slots=True)
60class StoredMemoryItemInfo(StoredDatastoreItemInfo):
61 """Internal InMemoryDatastore Metadata associated with a stored
62 DatasetRef.
63 """
65 timestamp: float
66 """Unix timestamp indicating the time the dataset was stored."""
68 storageClass: StorageClass
69 """StorageClass associated with the dataset."""
71 parentID: DatasetId
72 """ID of the parent `DatasetRef` if this entry is a concrete
73 composite. Not used if the dataset being stored is not a
74 virtual component of a composite
75 """
78class InMemoryDatastore(GenericBaseDatastore[StoredMemoryItemInfo]):
79 """Basic Datastore for writing to an in memory cache.
81 This datastore is ephemeral in that the contents of the datastore
82 disappear when the Python process completes. This also means that
83 other processes can not access this datastore.
85 Parameters
86 ----------
87 config : `DatastoreConfig` or `str`
88 Configuration.
89 bridgeManager : `DatastoreRegistryBridgeManager`
90 Object that manages the interface between `Registry` and datastores.
92 Notes
93 -----
94 InMemoryDatastore does not support any file-based ingest.
95 """
97 defaultConfigFile = "datastores/inMemoryDatastore.yaml"
98 """Path to configuration defaults. Accessed within the ``configs`` resource
99 or relative to a search path. Can be None if no defaults specified.
100 """
102 isEphemeral = True
103 """A new datastore is created every time and datasets disappear when
104 the process shuts down."""
106 datasets: dict[DatasetId, Any]
107 """Internal storage of datasets indexed by dataset ID."""
109 records: dict[DatasetId, StoredMemoryItemInfo]
110 """Internal records about stored datasets."""
112 def __init__(
113 self,
114 config: DatastoreConfig,
115 bridgeManager: DatastoreRegistryBridgeManager,
116 ):
117 super().__init__(config, bridgeManager)
119 # Name ourselves with the timestamp the datastore
120 # was created.
121 self.name = f"{type(self).__name__}@{time.time()}"
122 log.debug("Creating datastore %s", self.name)
124 # Storage of datasets, keyed by dataset_id
125 self.datasets: dict[DatasetId, Any] = {}
127 # Records is distinct in order to track concrete composite components
128 # where we register multiple components for a single dataset.
129 self.records: dict[DatasetId, StoredMemoryItemInfo] = {}
131 # Related records that share the same parent
132 self.related: dict[DatasetId, set[DatasetId]] = {}
134 self._trashedIds: set[DatasetId] = set()
136 @classmethod
137 def _create_from_config(
138 cls,
139 config: DatastoreConfig,
140 bridgeManager: DatastoreRegistryBridgeManager,
141 butlerRoot: ResourcePathExpression | None,
142 ) -> InMemoryDatastore:
143 return InMemoryDatastore(config, bridgeManager)
145 def clone(self, bridgeManager: DatastoreRegistryBridgeManager) -> InMemoryDatastore:
146 clone = InMemoryDatastore(self.config, bridgeManager)
147 # Sharing these objects is not thread-safe, but this class is only used
148 # in single-threaded test code.
149 clone.datasets = self.datasets
150 clone.records = self.records
151 clone.related = self.related
152 clone._trashedIds = self._trashedIds
153 return clone
155 @classmethod
156 def setConfigRoot(cls, root: str, config: Config, full: Config, overwrite: bool = True) -> None:
157 """Set any filesystem-dependent config options for this Datastore to
158 be appropriate for a new empty repository with the given root.
160 Does nothing in this implementation.
162 Parameters
163 ----------
164 root : `str`
165 Filesystem path to the root of the data repository.
166 config : `Config`
167 A `Config` to update. Only the subset understood by
168 this component will be updated. Will not expand
169 defaults.
170 full : `Config`
171 A complete config with all defaults expanded that can be
172 converted to a `DatastoreConfig`. Read-only and will not be
173 modified by this method.
174 Repository-specific options that should not be obtained
175 from defaults when Butler instances are constructed
176 should be copied from ``full`` to ``config``.
177 overwrite : `bool`, optional
178 If `False`, do not modify a value in ``config`` if the value
179 already exists. Default is always to overwrite with the provided
180 ``root``.
182 Notes
183 -----
184 If a keyword is explicitly defined in the supplied ``config`` it
185 will not be overridden by this method if ``overwrite`` is `False`.
186 This allows explicit values set in external configs to be retained.
187 """
188 return
190 def _get_stored_item_info(self, dataset_id: DatasetId) -> StoredMemoryItemInfo:
191 # Docstring inherited from GenericBaseDatastore.
192 return self.records[dataset_id]
194 def _remove_stored_item_info(self, dataset_id: DatasetId) -> None:
195 # Docstring inherited from GenericBaseDatastore.
196 # If a component has been removed previously then we can sometimes
197 # be asked to remove it again. Other datastores ignore this
198 # so also ignore here
199 if dataset_id not in self.records:
200 return
201 record = self.records[dataset_id]
202 del self.records[dataset_id]
203 self.related[record.parentID].remove(dataset_id)
205 def removeStoredItemInfo(self, ref: DatasetIdRef) -> None:
206 """Remove information about the file associated with this dataset.
208 Parameters
209 ----------
210 ref : `DatasetRef`
211 The dataset that has been removed.
213 Notes
214 -----
215 This method is actually not used by this implementation, but there are
216 some tests that check that this method works, so we keep it for now.
217 """
218 self._remove_stored_item_info(ref.id)
220 def _get_dataset_info(self, dataset_id: DatasetId) -> tuple[DatasetId, StoredMemoryItemInfo]:
221 """Check that the dataset is present and return the real ID and
222 associated information.
224 Parameters
225 ----------
226 dataset_id : `DatasetRef`
227 Target `DatasetRef`
229 Returns
230 -------
231 realID : `int`
232 The dataset ID associated with this ref that should be used. This
233 could either be the ID of the supplied `DatasetRef` or the parent.
234 storageInfo : `StoredMemoryItemInfo`
235 Associated storage information.
237 Raises
238 ------
239 FileNotFoundError
240 Raised if the dataset is not present in this datastore.
241 """
242 try:
243 storedItemInfo = self._get_stored_item_info(dataset_id)
244 except KeyError:
245 raise FileNotFoundError(f"No such file dataset in memory: {dataset_id}") from None
246 realID = dataset_id
247 if storedItemInfo.parentID is not None: 247 ↛ 250line 247 didn't jump to line 250 because the condition on line 247 was always true
248 realID = storedItemInfo.parentID
250 if realID not in self.datasets: 250 ↛ 251line 250 didn't jump to line 251 because the condition on line 250 was never true
251 raise FileNotFoundError(f"No such file dataset in memory: {dataset_id}")
253 return realID, storedItemInfo
255 def knows(self, ref: DatasetRef) -> bool:
256 """Check if the dataset is known to the datastore.
258 This datastore does not distinguish dataset existence from knowledge
259 of a dataset.
261 Parameters
262 ----------
263 ref : `DatasetRef`
264 Reference to the required dataset.
266 Returns
267 -------
268 exists : `bool`
269 `True` if the dataset is known to the datastore.
270 """
271 return self.exists(ref)
273 def exists(self, ref: DatasetRef) -> bool:
274 """Check if the dataset exists in the datastore.
276 Parameters
277 ----------
278 ref : `DatasetRef`
279 Reference to the required dataset.
281 Returns
282 -------
283 exists : `bool`
284 `True` if the entity exists in the `Datastore`.
285 """
286 try:
287 self._get_dataset_info(ref.id)
288 except FileNotFoundError:
289 return False
290 return True
292 def get(
293 self,
294 ref: DatasetRef,
295 parameters: Mapping[str, Any] | None = None,
296 storageClass: StorageClass | str | None = None,
297 ) -> Any:
298 """Load an InMemoryDataset from the store.
300 Parameters
301 ----------
302 ref : `DatasetRef`
303 Reference to the required Dataset.
304 parameters : `dict`
305 `StorageClass`-specific parameters that specify, for example,
306 a slice of the dataset to be loaded.
307 storageClass : `StorageClass` or `str`, optional
308 The storage class to be used to override the Python type
309 returned by this method. By default the returned type matches
310 the dataset type definition for this dataset. Specifying a
311 read `StorageClass` can force a different type to be returned.
312 This type must be compatible with the original type.
314 Returns
315 -------
316 inMemoryDataset : `object`
317 Requested dataset or slice thereof as an InMemoryDataset.
319 Raises
320 ------
321 FileNotFoundError
322 Requested dataset can not be retrieved.
323 TypeError
324 Return value from formatter has unexpected type.
325 ValueError
326 Formatter failed to process the dataset.
327 """
328 log.debug("Retrieve %s from %s with parameters %s", ref, self.name, parameters)
330 realID, storedItemInfo = self._get_dataset_info(ref.id)
332 # We have a write storage class and a read storage class and they
333 # can be different for concrete composites or if overridden.
334 if storageClass is not None:
335 ref = ref.overrideStorageClass(storageClass)
336 refStorageClass = ref.datasetType.storageClass
337 writeStorageClass = storedItemInfo.storageClass
339 component = ref.datasetType.component()
341 # Check that the supplied parameters are suitable for the type read
342 # If this is a derived component we validate against the composite
343 isDerivedComponent = False
344 if component in writeStorageClass.derivedComponents:
345 writeStorageClass.validateParameters(parameters)
346 isDerivedComponent = True
347 else:
348 refStorageClass.validateParameters(parameters)
350 inMemoryDataset = self.datasets[realID]
352 # if this is a read only component we need to apply parameters
353 # before we retrieve the component. We assume that the parameters
354 # will affect the data globally, before the derived component
355 # is selected.
356 if isDerivedComponent:
357 inMemoryDataset = writeStorageClass.delegate().handleParameters(inMemoryDataset, parameters)
358 # Then disable parameters for later
359 parameters = {}
361 # Check if we have a component.
362 if component:
363 # In-memory datastore must have stored the dataset as a single
364 # object in the write storage class. We therefore use that
365 # storage class delegate to obtain the component.
366 compositeStorageClass = writeStorageClass
367 if component not in writeStorageClass.allComponents():
368 # The component is defined by the storage class the caller is
369 # converting to rather than the one used to write, so the
370 # composite has to be converted before the component can be
371 # extracted from it.
372 parentStorageClass = ref.datasetType.parentStorageClass
373 if parentStorageClass is not None: 373 ↛ 376line 373 didn't jump to line 376 because the condition on line 373 was always true
374 compositeStorageClass = parentStorageClass
375 inMemoryDataset = parentStorageClass.coerce_type(inMemoryDataset)
376 inMemoryDataset = compositeStorageClass.delegate().getComponent(inMemoryDataset, component)
378 # Since there is no formatter to process parameters, they all must be
379 # passed to the assembler.
380 inMemoryDataset = post_process_get(
381 inMemoryDataset, refStorageClass, parameters, isComponent=component is not None
382 )
384 # Last minute type conversion.
385 return refStorageClass.coerce_type(inMemoryDataset)
387 def put(self, inMemoryDataset: Any, ref: DatasetRef, provenance: DatasetProvenance | None = None) -> None:
388 """Write a InMemoryDataset with a given `DatasetRef` to the store.
390 Parameters
391 ----------
392 inMemoryDataset : `object`
393 The dataset to store.
394 ref : `DatasetRef`
395 Reference to the associated Dataset.
396 provenance : `DatasetProvenance` or `None`, optional
397 Any provenance that should be attached to the serialized dataset.
399 Raises
400 ------
401 TypeError
402 Supplied object and storage class are inconsistent.
403 DatasetTypeNotSupportedError
404 The associated `DatasetType` is not handled by this datastore.
406 Notes
407 -----
408 If the datastore is configured to reject certain dataset types it
409 is possible that the put will fail and raise a
410 `DatasetTypeNotSupportedError`. The main use case for this is to
411 allow `ChainedDatastore` to put to multiple datastores without
412 requiring that every datastore accepts the dataset.
413 """
414 if not self.constraints.isAcceptable(ref):
415 # Raise rather than use boolean return value.
416 raise DatasetTypeNotSupportedError(
417 f"Dataset {ref} has been rejected by this datastore via configuration."
418 )
420 # Do not override existing datasets.
421 if ref.id in self.records:
422 raise DuplicateIdError(f"Attempt to override existing dataset: {ref.id}")
424 # May need to coerce the in memory dataset to the correct
425 # python type, otherwise parameters may not work.
426 try:
427 delegate = ref.datasetType.storageClass.delegate()
428 except TypeError:
429 # TypeError is raised when a storage class doesn't have a delegate.
430 delegate = None
431 if not delegate or not delegate.can_accept(inMemoryDataset):
432 inMemoryDataset = ref.datasetType.storageClass.coerce_type(inMemoryDataset)
434 # Update provenance.
435 if delegate: 435 ↛ 438line 435 didn't jump to line 438 because the condition on line 435 was always true
436 inMemoryDataset = delegate.add_provenance(inMemoryDataset, ref, provenance=provenance)
438 self.datasets[ref.id] = inMemoryDataset
439 log.debug("Store %s in %s", ref, self.name)
441 # Store time we received this content, to allow us to optionally
442 # expire it. Instead of storing a filename here, we include the
443 # ID of this datasetRef so we can find it from components.
444 itemInfo = StoredMemoryItemInfo(time.time(), ref.datasetType.storageClass, parentID=ref.id)
446 # We have to register this content with registry.
447 # Currently this assumes we have a file so we need to use stub entries
448 self.records[ref.id] = itemInfo
449 self.related.setdefault(itemInfo.parentID, set()).add(ref.id)
451 if self._transaction is not None:
452 self._transaction.registerUndo("put", self.remove, ref)
454 def getURIs(self, ref: DatasetRef, predict: bool = False) -> DatasetRefURIs:
455 """Return URIs associated with dataset.
457 Parameters
458 ----------
459 ref : `DatasetRef`
460 Reference to the required dataset.
461 predict : `bool`, optional
462 If the datastore does not know about the dataset, controls whether
463 it should return a predicted URI or not.
465 Returns
466 -------
467 uris : `DatasetRefURIs`
468 The URI to the primary artifact associated with this dataset (if
469 the dataset was disassembled within the datastore this may be
470 `None`), and the URIs to any components associated with the dataset
471 artifact. (can be empty if there are no components).
473 Notes
474 -----
475 The URIs returned for in-memory datastores are not usable but
476 provide an indication of the associated dataset.
477 """
478 # Include the dataID as a URI query
479 query = urlencode(ref.dataId.required)
481 # if this has never been written then we have to guess
482 if not self.exists(ref):
483 if not predict:
484 raise FileNotFoundError(f"Dataset {ref} not in this datastore")
485 name = f"{ref.datasetType.name}"
486 fragment = "#predicted"
487 else:
488 realID, _ = self._get_dataset_info(ref.id)
489 name = f"{id(self.datasets[realID])}_{ref.id}"
490 fragment = ""
492 return DatasetRefURIs(ResourcePath(f"mem://{name}?{query}{fragment}"), {})
494 def getURI(self, ref: DatasetRef, predict: bool = False) -> ResourcePath:
495 """URI to the Dataset.
497 Always uses "mem://" URI prefix.
499 Parameters
500 ----------
501 ref : `DatasetRef`
502 Reference to the required Dataset.
503 predict : `bool`
504 If `True`, allow URIs to be returned of datasets that have not
505 been written.
507 Returns
508 -------
509 uri : `str`
510 URI pointing to the dataset within the datastore. If the
511 dataset does not exist in the datastore, and if ``predict`` is
512 `True`, the URI will be a prediction and will include a URI
513 fragment "#predicted".
514 If the datastore does not have entities that relate well
515 to the concept of a URI the returned URI string will be
516 descriptive. The returned URI is not guaranteed to be obtainable.
518 Raises
519 ------
520 FileNotFoundError
521 A URI has been requested for a dataset that does not exist and
522 guessing is not allowed.
523 AssertionError
524 Raised if an internal error occurs.
525 """
526 primary, _ = self.getURIs(ref, predict)
527 if primary is None:
528 # This should be impossible since this datastore does
529 # not disassemble. This check also helps mypy.
530 raise AssertionError(f"Unexpectedly got no URI for in-memory datastore for {ref}")
531 return primary
533 def ingest_zip(self, zip_path: ResourcePath, transfer: str | None, *, dry_run: bool = False) -> None:
534 raise NotImplementedError("Can only ingest a Zip into a file datastore.")
536 def retrieveArtifacts(
537 self,
538 refs: Iterable[DatasetRef],
539 destination: ResourcePath,
540 transfer: str = "auto",
541 preserve_path: bool = True,
542 overwrite: bool | None = False,
543 write_index: bool = True,
544 add_prefix: bool = False,
545 ) -> dict[ResourcePath, ArtifactIndexInfo]:
546 """Retrieve the file artifacts associated with the supplied refs.
548 Parameters
549 ----------
550 refs : `~collections.abc.Iterable` of `DatasetRef`
551 The datasets for which artifacts are to be retrieved.
552 A single ref can result in multiple artifacts. The refs must
553 be resolved.
554 destination : `lsst.resources.ResourcePath`
555 Location to write the artifacts.
556 transfer : `str`, optional
557 Method to use to transfer the artifacts. Must be one of the options
558 supported by `lsst.resources.ResourcePath.transfer_from`.
559 "move" is not allowed.
560 preserve_path : `bool`, optional
561 If `True` the full path of the artifact within the datastore
562 is preserved. If `False` the final file component of the path
563 is used.
564 overwrite : `bool`, optional
565 If `True` allow transfers to overwrite existing files at the
566 destination.
567 write_index : `bool`, optional
568 If `True` write a file at the top level containing a serialization
569 of a `ZipIndex` for the downloaded datasets.
570 add_prefix : `bool`, optional
571 If `True` and if ``preserve_path`` is `False`, apply a prefix to
572 the filenames corresponding to some part of the dataset ref ID.
573 This can be used to guarantee uniqueness.
575 Notes
576 -----
577 Not implemented by this datastore.
578 """
579 # Could conceivably launch a FileDatastore to use formatters to write
580 # the data but this is fraught with problems.
581 raise NotImplementedError("Can not write artifacts to disk from in-memory datastore.")
583 def forget(self, refs: Iterable[DatasetRef]) -> None:
584 # Docstring inherited.
585 refs = list(refs)
586 for ref in refs:
587 self._remove_stored_item_info(ref.id)
589 @transactional
590 def trash(self, ref: DatasetRef | Iterable[DatasetRef], ignore_errors: bool = False) -> None:
591 """Indicate to the Datastore that a dataset can be removed.
593 Parameters
594 ----------
595 ref : `DatasetRef` or iterable thereof
596 Reference to the required Dataset(s).
597 ignore_errors : `bool`, optional
598 Indicate that errors should be ignored.
600 Returns
601 -------
602 None
604 Raises
605 ------
606 FileNotFoundError
607 Attempt to remove a dataset that does not exist. Only relevant
608 if a single dataset ref is given.
610 Notes
611 -----
612 Concurrency should not normally be an issue for the in memory datastore
613 since all internal changes are isolated to solely this process and
614 the registry only changes rows associated with this process.
615 """
616 if isinstance(ref, DatasetRef):
617 # Check that this dataset is known to datastore
618 try:
619 self._get_dataset_info(ref.id)
620 except Exception as e:
621 if ignore_errors: 621 ↛ 622line 621 didn't jump to line 622 because the condition on line 621 was never true
622 log.warning(
623 "Error encountered moving dataset %s to trash in datastore %s: %s", ref, self.name, e
624 )
625 else:
626 raise
627 log.debug("Trash %s in datastore %s", ref, self.name)
628 ref_list = [ref]
629 else:
630 ref_list = list(ref)
631 log.debug("Bulk trashing of datasets in datastore %s", self.name)
633 def _rollbackMoveToTrash(refs: Iterable[DatasetIdRef]) -> None:
634 for ref in refs: 634 ↛ exitline 634 didn't return from function '_rollbackMoveToTrash' because the loop on line 634 didn't complete
635 self._trashedIds.remove(ref.id)
637 assert self._transaction is not None, "Must be in transaction"
638 with self._transaction.undoWith(f"Trash {len(ref_list)} datasets", _rollbackMoveToTrash, ref_list):
639 self._trashedIds.update(ref.id for ref in ref_list)
641 def emptyTrash(
642 self, ignore_errors: bool = False, refs: Collection[DatasetRef] | None = None, dry_run: bool = False
643 ) -> set[ResourcePath]:
644 """Remove all datasets from the trash.
646 Parameters
647 ----------
648 ignore_errors : `bool`, optional
649 Ignore errors.
650 refs : `collections.abc.Collection` [ `DatasetRef` ] or `None`
651 Explicit list of datasets that can be removed from trash. If listed
652 datasets are not already stored in the trash table they will be
653 ignored. If `None` every entry in the trash table will be
654 processed.
655 dry_run : `bool`, optional
656 If `True`, the trash table will be queried and results reported
657 but no artifacts will be removed.
659 Returns
660 -------
661 removed : `set` [ `lsst.resources.ResourcePath` ]
662 List of artifacts that were removed. Empty for this datastore.
664 Notes
665 -----
666 The internal tracking of datasets is affected by this method and
667 transaction handling is not supported if there is a problem before
668 the datasets themselves are deleted.
670 Concurrency should not normally be an issue for the in memory datastore
671 since all internal changes are isolated to solely this process and
672 the registry only changes rows associated with this process.
673 """
674 log.debug("Emptying trash in datastore %s", self.name)
676 trashed_ids = self._trashedIds
677 if refs: 677 ↛ 682line 677 didn't jump to line 682 because the condition on line 677 was always true
678 selected_ids = {ref.id for ref in refs}
679 if selected_ids is not None: 679 ↛ 682line 679 didn't jump to line 682 because the condition on line 679 was always true
680 trashed_ids = {tid for tid in trashed_ids if tid in selected_ids}
682 if dry_run: 682 ↛ 683line 682 didn't jump to line 683 because the condition on line 682 was never true
683 log.info(
684 "Would attempt remove %s dataset%s.", len(trashed_ids), "s" if len(trashed_ids) != 1 else ""
685 )
686 return set()
688 for dataset_id in trashed_ids:
689 try:
690 realID, _ = self._get_dataset_info(dataset_id)
691 except FileNotFoundError:
692 # Dataset already removed so ignore it
693 continue
694 except Exception as e:
695 if ignore_errors:
696 log.warning(
697 "Emptying trash in datastore %s but encountered an error with dataset %s: %s",
698 self.name,
699 dataset_id,
700 e,
701 )
702 continue
703 else:
704 raise
706 # Determine whether all references to this dataset have been
707 # removed and we can delete the dataset itself
708 allRefs = self.related[realID]
709 remainingRefs = allRefs - {dataset_id}
710 if not remainingRefs: 710 ↛ 715line 710 didn't jump to line 715 because the condition on line 710 was always true
711 log.debug("Removing artifact %s from datastore %s", realID, self.name)
712 del self.datasets[realID]
714 # Remove this entry
715 self._remove_stored_item_info(dataset_id)
717 # Empty the trash table
718 self._trashedIds = self._trashedIds - trashed_ids
719 return set()
721 def validateConfiguration(
722 self, entities: Iterable[DatasetRef | DatasetType | StorageClass], logFailures: bool = False
723 ) -> None:
724 """Validate some of the configuration for this datastore.
726 Parameters
727 ----------
728 entities : `~collections.abc.Iterable` [`DatasetRef` | `DatasetType`\
729 | `StorageClass`]
730 Entities to test against this configuration. Can be differing
731 types.
732 logFailures : `bool`, optional
733 If `True`, output a log message for every validation error
734 detected.
736 Returns
737 -------
738 None
740 Raises
741 ------
742 DatastoreValidationError
743 Raised if there is a validation problem with a configuration.
744 All the problems are reported in a single exception.
746 Notes
747 -----
748 This method is a no-op.
749 """
750 return
752 def _overrideTransferMode(self, *datasets: Any, transfer: str | None = None) -> str | None:
753 # Docstring is inherited from base class
754 return transfer
756 def validateKey(self, lookupKey: LookupKey, entity: DatasetRef | DatasetType | StorageClass) -> None:
757 # Docstring is inherited from base class
758 return
760 def getLookupKeys(self) -> set[LookupKey]:
761 # Docstring is inherited from base class
762 return self.constraints.getLookupKeys()
764 def needs_expanded_data_ids(
765 self,
766 transfer: str | None,
767 entity: DatasetRef | DatasetType | StorageClass | None = None,
768 ) -> bool:
769 # Docstring inherited.
770 return False
772 def import_records(self, data: Mapping[str, DatastoreRecordData]) -> None:
773 # Docstring inherited from the base class.
774 return
776 def export_records(self, refs: Iterable[DatasetIdRef]) -> Mapping[str, DatastoreRecordData]:
777 # Docstring inherited from the base class.
779 # In-memory Datastore records cannot be exported or imported
780 return {}
782 def export_predicted_records(self, refs: Iterable[DatasetIdRef]) -> dict[str, DatastoreRecordData]:
783 # Docstring inherited from the base class.
785 # In-memory Datastore records cannot be exported or imported
786 return {}
788 def get_opaque_table_definitions(self) -> Mapping[str, DatastoreOpaqueTable]:
789 # Docstring inherited from the base class.
790 return {}