Coverage for python/lsst/daf/butler/registry/_registry.py: 78%
147 statements
« prev ^ index » next coverage.py v7.2.5, created at 2023-05-18 09:13 +0000
« prev ^ index » next coverage.py v7.2.5, created at 2023-05-18 09:13 +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 program is free software: you can redistribute it and/or modify
10# it under the terms of the GNU General Public License as published by
11# the Free Software Foundation, either version 3 of the License, or
12# (at your option) any later version.
13#
14# This program is distributed in the hope that it will be useful,
15# but WITHOUT ANY WARRANTY; without even the implied warranty of
16# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
17# GNU General Public License for more details.
18#
19# You should have received a copy of the GNU General Public License
20# along with this program. If not, see <http://www.gnu.org/licenses/>.
22from __future__ import annotations
24__all__ = ("Registry",)
26import contextlib
27import logging
28import re
29from abc import ABC, abstractmethod
30from typing import (
31 TYPE_CHECKING,
32 Any,
33 Dict,
34 Iterable,
35 Iterator,
36 List,
37 Mapping,
38 Optional,
39 Sequence,
40 Set,
41 Tuple,
42 Type,
43 Union,
44)
46from lsst.resources import ResourcePathExpression
47from lsst.utils import doImportType
48from lsst.utils.ellipsis import Ellipsis, EllipsisType
50from ..core import (
51 Config,
52 DataCoordinate,
53 DataId,
54 DatasetAssociation,
55 DatasetId,
56 DatasetIdFactory,
57 DatasetIdGenEnum,
58 DatasetRef,
59 DatasetType,
60 Dimension,
61 DimensionConfig,
62 DimensionElement,
63 DimensionGraph,
64 DimensionRecord,
65 DimensionUniverse,
66 NameLookupMapping,
67 StorageClassFactory,
68 Timespan,
69)
70from ._collection_summary import CollectionSummary
71from ._collectionType import CollectionType
72from ._config import RegistryConfig
73from ._defaults import RegistryDefaults
74from .queries import DataCoordinateQueryResults, DatasetQueryResults, DimensionRecordQueryResults
75from .wildcards import CollectionWildcard
77if TYPE_CHECKING:
78 from .._butlerConfig import ButlerConfig
79 from .interfaces import CollectionRecord, DatastoreRegistryBridgeManager, ObsCoreTableManager
81_LOG = logging.getLogger(__name__)
83# TYpe alias for `collections` arguments.
84CollectionArgType = str | re.Pattern | Iterable[str | re.Pattern] | EllipsisType | CollectionWildcard
87class Registry(ABC):
88 """Abstract Registry interface.
90 Each registry implementation can have its own constructor parameters.
91 The assumption is that an instance of a specific subclass will be
92 constructed from configuration using `Registry.fromConfig()`.
93 The base class will look for a ``cls`` entry and call that specific
94 `fromConfig()` method.
96 All subclasses should store `RegistryDefaults` in a ``_defaults``
97 property. No other properties are assumed shared between implementations.
98 """
100 defaultConfigFile: Optional[str] = None
101 """Path to configuration defaults. Accessed within the ``configs`` resource
102 or relative to a search path. Can be None if no defaults specified.
103 """
105 @classmethod
106 def forceRegistryConfig(
107 cls, config: Optional[Union[ButlerConfig, RegistryConfig, Config, str]]
108 ) -> RegistryConfig:
109 """Force the supplied config to a `RegistryConfig`.
111 Parameters
112 ----------
113 config : `RegistryConfig`, `Config` or `str` or `None`
114 Registry configuration, if missing then default configuration will
115 be loaded from registry.yaml.
117 Returns
118 -------
119 registry_config : `RegistryConfig`
120 A registry config.
121 """
122 if not isinstance(config, RegistryConfig):
123 if isinstance(config, (str, Config)) or config is None:
124 config = RegistryConfig(config)
125 else:
126 raise ValueError(f"Incompatible Registry configuration: {config}")
127 return config
129 @classmethod
130 def determineTrampoline(
131 cls, config: Optional[Union[ButlerConfig, RegistryConfig, Config, str]]
132 ) -> Tuple[Type[Registry], RegistryConfig]:
133 """Return class to use to instantiate real registry.
135 Parameters
136 ----------
137 config : `RegistryConfig` or `str`, optional
138 Registry configuration, if missing then default configuration will
139 be loaded from registry.yaml.
141 Returns
142 -------
143 requested_cls : `type` of `Registry`
144 The real registry class to use.
145 registry_config : `RegistryConfig`
146 The `RegistryConfig` to use.
147 """
148 config = cls.forceRegistryConfig(config)
150 # Default to the standard registry
151 registry_cls_name = config.get("cls", "lsst.daf.butler.registries.sql.SqlRegistry")
152 registry_cls = doImportType(registry_cls_name)
153 if registry_cls is cls:
154 raise ValueError("Can not instantiate the abstract base Registry from config")
155 if not issubclass(registry_cls, Registry):
156 raise TypeError(
157 f"Registry class obtained from config {registry_cls_name} is not a Registry class."
158 )
159 return registry_cls, config
161 @classmethod
162 def createFromConfig(
163 cls,
164 config: Optional[Union[RegistryConfig, str]] = None,
165 dimensionConfig: Optional[Union[DimensionConfig, str]] = None,
166 butlerRoot: Optional[ResourcePathExpression] = None,
167 ) -> Registry:
168 """Create registry database and return `Registry` instance.
170 This method initializes database contents, database must be empty
171 prior to calling this method.
173 Parameters
174 ----------
175 config : `RegistryConfig` or `str`, optional
176 Registry configuration, if missing then default configuration will
177 be loaded from registry.yaml.
178 dimensionConfig : `DimensionConfig` or `str`, optional
179 Dimensions configuration, if missing then default configuration
180 will be loaded from dimensions.yaml.
181 butlerRoot : convertible to `lsst.resources.ResourcePath`, optional
182 Path to the repository root this `Registry` will manage.
184 Returns
185 -------
186 registry : `Registry`
187 A new `Registry` instance.
189 Notes
190 -----
191 This class will determine the concrete `Registry` subclass to
192 use from configuration. Each subclass should implement this method
193 even if it can not create a registry.
194 """
195 registry_cls, registry_config = cls.determineTrampoline(config)
196 return registry_cls.createFromConfig(registry_config, dimensionConfig, butlerRoot)
198 @classmethod
199 def fromConfig(
200 cls,
201 config: Union[ButlerConfig, RegistryConfig, Config, str],
202 butlerRoot: Optional[ResourcePathExpression] = None,
203 writeable: bool = True,
204 defaults: Optional[RegistryDefaults] = None,
205 ) -> Registry:
206 """Create `Registry` subclass instance from `config`.
208 Registry database must be initialized prior to calling this method.
210 Parameters
211 ----------
212 config : `ButlerConfig`, `RegistryConfig`, `Config` or `str`
213 Registry configuration
214 butlerRoot : `lsst.resources.ResourcePathExpression`, optional
215 Path to the repository root this `Registry` will manage.
216 writeable : `bool`, optional
217 If `True` (default) create a read-write connection to the database.
218 defaults : `RegistryDefaults`, optional
219 Default collection search path and/or output `~CollectionType.RUN`
220 collection.
222 Returns
223 -------
224 registry : `Registry` (subclass)
225 A new `Registry` subclass instance.
227 Notes
228 -----
229 This class will determine the concrete `Registry` subclass to
230 use from configuration. Each subclass should implement this method.
231 """
232 # The base class implementation should trampoline to the correct
233 # subclass. No implementation should ever use this implementation
234 # directly. If no class is specified, default to the standard
235 # registry.
236 registry_cls, registry_config = cls.determineTrampoline(config)
237 return registry_cls.fromConfig(config, butlerRoot, writeable, defaults)
239 @abstractmethod
240 def isWriteable(self) -> bool:
241 """Return `True` if this registry allows write operations, and `False`
242 otherwise.
243 """
244 raise NotImplementedError()
246 @abstractmethod
247 def copy(self, defaults: Optional[RegistryDefaults] = None) -> Registry:
248 """Create a new `Registry` backed by the same data repository and
249 connection as this one, but independent defaults.
251 Parameters
252 ----------
253 defaults : `RegistryDefaults`, optional
254 Default collections and data ID values for the new registry. If
255 not provided, ``self.defaults`` will be used (but future changes
256 to either registry's defaults will not affect the other).
258 Returns
259 -------
260 copy : `Registry`
261 A new `Registry` instance with its own defaults.
263 Notes
264 -----
265 Because the new registry shares a connection with the original, they
266 also share transaction state (despite the fact that their `transaction`
267 context manager methods do not reflect this), and must be used with
268 care.
269 """
270 raise NotImplementedError()
272 @property
273 @abstractmethod
274 def dimensions(self) -> DimensionUniverse:
275 """Definitions of all dimensions recognized by this `Registry`
276 (`DimensionUniverse`).
277 """
278 raise NotImplementedError()
280 @property
281 def defaults(self) -> RegistryDefaults:
282 """Default collection search path and/or output `~CollectionType.RUN`
283 collection (`RegistryDefaults`).
285 This is an immutable struct whose components may not be set
286 individually, but the entire struct can be set by assigning to this
287 property.
288 """
289 return self._defaults
291 @defaults.setter
292 def defaults(self, value: RegistryDefaults) -> None:
293 if value.run is not None:
294 self.registerRun(value.run)
295 value.finish(self)
296 self._defaults = value
298 @abstractmethod
299 def refresh(self) -> None:
300 """Refresh all in-memory state by querying the database.
302 This may be necessary to enable querying for entities added by other
303 registry instances after this one was constructed.
304 """
305 raise NotImplementedError()
307 @contextlib.contextmanager
308 @abstractmethod
309 def transaction(self, *, savepoint: bool = False) -> Iterator[None]:
310 """Return a context manager that represents a transaction."""
311 raise NotImplementedError()
313 def resetConnectionPool(self) -> None:
314 """Reset connection pool for registry if relevant.
316 This operation can be used reset connections to servers when
317 using registry with fork-based multiprocessing. This method should
318 usually be called by the child process immediately
319 after the fork.
321 The base class implementation is a no-op.
322 """
323 pass
325 @abstractmethod
326 def registerCollection(
327 self, name: str, type: CollectionType = CollectionType.TAGGED, doc: Optional[str] = None
328 ) -> bool:
329 """Add a new collection if one with the given name does not exist.
331 Parameters
332 ----------
333 name : `str`
334 The name of the collection to create.
335 type : `CollectionType`
336 Enum value indicating the type of collection to create.
337 doc : `str`, optional
338 Documentation string for the collection.
340 Returns
341 -------
342 registered : `bool`
343 Boolean indicating whether the collection was already registered
344 or was created by this call.
346 Notes
347 -----
348 This method cannot be called within transactions, as it needs to be
349 able to perform its own transaction to be concurrent.
350 """
351 raise NotImplementedError()
353 @abstractmethod
354 def getCollectionType(self, name: str) -> CollectionType:
355 """Return an enumeration value indicating the type of the given
356 collection.
358 Parameters
359 ----------
360 name : `str`
361 The name of the collection.
363 Returns
364 -------
365 type : `CollectionType`
366 Enum value indicating the type of this collection.
368 Raises
369 ------
370 MissingCollectionError
371 Raised if no collection with the given name exists.
372 """
373 raise NotImplementedError()
375 @abstractmethod
376 def _get_collection_record(self, name: str) -> CollectionRecord:
377 """Return the record for this collection.
379 Parameters
380 ----------
381 name : `str`
382 Name of the collection for which the record is to be retrieved.
384 Returns
385 -------
386 record : `CollectionRecord`
387 The record for this collection.
388 """
389 raise NotImplementedError()
391 @abstractmethod
392 def registerRun(self, name: str, doc: Optional[str] = None) -> bool:
393 """Add a new run if one with the given name does not exist.
395 Parameters
396 ----------
397 name : `str`
398 The name of the run to create.
399 doc : `str`, optional
400 Documentation string for the collection.
402 Returns
403 -------
404 registered : `bool`
405 Boolean indicating whether a new run was registered. `False`
406 if it already existed.
408 Notes
409 -----
410 This method cannot be called within transactions, as it needs to be
411 able to perform its own transaction to be concurrent.
412 """
413 raise NotImplementedError()
415 @abstractmethod
416 def removeCollection(self, name: str) -> None:
417 """Remove the given collection from the registry.
419 Parameters
420 ----------
421 name : `str`
422 The name of the collection to remove.
424 Raises
425 ------
426 MissingCollectionError
427 Raised if no collection with the given name exists.
428 sqlalchemy.IntegrityError
429 Raised if the database rows associated with the collection are
430 still referenced by some other table, such as a dataset in a
431 datastore (for `~CollectionType.RUN` collections only) or a
432 `~CollectionType.CHAINED` collection of which this collection is
433 a child.
435 Notes
436 -----
437 If this is a `~CollectionType.RUN` collection, all datasets and quanta
438 in it will removed from the `Registry` database. This requires that
439 those datasets be removed (or at least trashed) from any datastores
440 that hold them first.
442 A collection may not be deleted as long as it is referenced by a
443 `~CollectionType.CHAINED` collection; the ``CHAINED`` collection must
444 be deleted or redefined first.
445 """
446 raise NotImplementedError()
448 @abstractmethod
449 def getCollectionChain(self, parent: str) -> Sequence[str]:
450 """Return the child collections in a `~CollectionType.CHAINED`
451 collection.
453 Parameters
454 ----------
455 parent : `str`
456 Name of the chained collection. Must have already been added via
457 a call to `Registry.registerCollection`.
459 Returns
460 -------
461 children : `Sequence` [ `str` ]
462 An ordered sequence of collection names that are searched when the
463 given chained collection is searched.
465 Raises
466 ------
467 MissingCollectionError
468 Raised if ``parent`` does not exist in the `Registry`.
469 CollectionTypeError
470 Raised if ``parent`` does not correspond to a
471 `~CollectionType.CHAINED` collection.
472 """
473 raise NotImplementedError()
475 @abstractmethod
476 def setCollectionChain(self, parent: str, children: Any, *, flatten: bool = False) -> None:
477 """Define or redefine a `~CollectionType.CHAINED` collection.
479 Parameters
480 ----------
481 parent : `str`
482 Name of the chained collection. Must have already been added via
483 a call to `Registry.registerCollection`.
484 children : `Any`
485 An expression defining an ordered search of child collections,
486 generally an iterable of `str`; see
487 :ref:`daf_butler_collection_expressions` for more information.
488 flatten : `bool`, optional
489 If `True` (`False` is default), recursively flatten out any nested
490 `~CollectionType.CHAINED` collections in ``children`` first.
492 Raises
493 ------
494 MissingCollectionError
495 Raised when any of the given collections do not exist in the
496 `Registry`.
497 CollectionTypeError
498 Raised if ``parent`` does not correspond to a
499 `~CollectionType.CHAINED` collection.
500 ValueError
501 Raised if the given collections contains a cycle.
502 """
503 raise NotImplementedError()
505 @abstractmethod
506 def getCollectionParentChains(self, collection: str) -> Set[str]:
507 """Return the CHAINED collections that directly contain the given one.
509 Parameters
510 ----------
511 name : `str`
512 Name of the collection.
514 Returns
515 -------
516 chains : `set` of `str`
517 Set of `~CollectionType.CHAINED` collection names.
518 """
519 raise NotImplementedError()
521 @abstractmethod
522 def getCollectionDocumentation(self, collection: str) -> Optional[str]:
523 """Retrieve the documentation string for a collection.
525 Parameters
526 ----------
527 name : `str`
528 Name of the collection.
530 Returns
531 -------
532 docs : `str` or `None`
533 Docstring for the collection with the given name.
534 """
535 raise NotImplementedError()
537 @abstractmethod
538 def setCollectionDocumentation(self, collection: str, doc: Optional[str]) -> None:
539 """Set the documentation string for a collection.
541 Parameters
542 ----------
543 name : `str`
544 Name of the collection.
545 docs : `str` or `None`
546 Docstring for the collection with the given name; will replace any
547 existing docstring. Passing `None` will remove any existing
548 docstring.
549 """
550 raise NotImplementedError()
552 @abstractmethod
553 def getCollectionSummary(self, collection: str) -> CollectionSummary:
554 """Return a summary for the given collection.
556 Parameters
557 ----------
558 collection : `str`
559 Name of the collection for which a summary is to be retrieved.
561 Returns
562 -------
563 summary : `CollectionSummary`
564 Summary of the dataset types and governor dimension values in
565 this collection.
566 """
567 raise NotImplementedError()
569 @abstractmethod
570 def registerDatasetType(self, datasetType: DatasetType) -> bool:
571 """
572 Add a new `DatasetType` to the Registry.
574 It is not an error to register the same `DatasetType` twice.
576 Parameters
577 ----------
578 datasetType : `DatasetType`
579 The `DatasetType` to be added.
581 Returns
582 -------
583 inserted : `bool`
584 `True` if ``datasetType`` was inserted, `False` if an identical
585 existing `DatsetType` was found. Note that in either case the
586 DatasetType is guaranteed to be defined in the Registry
587 consistently with the given definition.
589 Raises
590 ------
591 ValueError
592 Raised if the dimensions or storage class are invalid.
593 ConflictingDefinitionError
594 Raised if this DatasetType is already registered with a different
595 definition.
597 Notes
598 -----
599 This method cannot be called within transactions, as it needs to be
600 able to perform its own transaction to be concurrent.
601 """
602 raise NotImplementedError()
604 @abstractmethod
605 def removeDatasetType(self, name: str | tuple[str, ...]) -> None:
606 """Remove the named `DatasetType` from the registry.
608 .. warning::
610 Registry implementations can cache the dataset type definitions.
611 This means that deleting the dataset type definition may result in
612 unexpected behavior from other butler processes that are active
613 that have not seen the deletion.
615 Parameters
616 ----------
617 name : `str` or `tuple[str, ...]`
618 Name of the type to be removed or tuple containing a list of type
619 names to be removed. Wildcards are allowed.
621 Raises
622 ------
623 lsst.daf.butler.registry.OrphanedRecordError
624 Raised if an attempt is made to remove the dataset type definition
625 when there are already datasets associated with it.
627 Notes
628 -----
629 If the dataset type is not registered the method will return without
630 action.
631 """
632 raise NotImplementedError()
634 @abstractmethod
635 def getDatasetType(self, name: str) -> DatasetType:
636 """Get the `DatasetType`.
638 Parameters
639 ----------
640 name : `str`
641 Name of the type.
643 Returns
644 -------
645 type : `DatasetType`
646 The `DatasetType` associated with the given name.
648 Raises
649 ------
650 MissingDatasetTypeError
651 Raised if the requested dataset type has not been registered.
653 Notes
654 -----
655 This method handles component dataset types automatically, though most
656 other registry operations do not.
657 """
658 raise NotImplementedError()
660 @abstractmethod
661 def supportsIdGenerationMode(self, mode: DatasetIdGenEnum) -> bool:
662 """Test whether the given dataset ID generation mode is supported by
663 `insertDatasets`.
665 Parameters
666 ----------
667 mode : `DatasetIdGenEnum`
668 Enum value for the mode to test.
670 Returns
671 -------
672 supported : `bool`
673 Whether the given mode is supported.
674 """
675 raise NotImplementedError()
677 @abstractmethod
678 def findDataset(
679 self,
680 datasetType: Union[DatasetType, str],
681 dataId: Optional[DataId] = None,
682 *,
683 collections: CollectionArgType | None = None,
684 timespan: Optional[Timespan] = None,
685 **kwargs: Any,
686 ) -> Optional[DatasetRef]:
687 """Find a dataset given its `DatasetType` and data ID.
689 This can be used to obtain a `DatasetRef` that permits the dataset to
690 be read from a `Datastore`. If the dataset is a component and can not
691 be found using the provided dataset type, a dataset ref for the parent
692 will be returned instead but with the correct dataset type.
694 Parameters
695 ----------
696 datasetType : `DatasetType` or `str`
697 A `DatasetType` or the name of one. If this is a `DatasetType`
698 instance, its storage class will be respected and propagated to
699 the output, even if it differs from the dataset type definition
700 in the registry, as long as the storage classes are convertible.
701 dataId : `dict` or `DataCoordinate`, optional
702 A `dict`-like object containing the `Dimension` links that identify
703 the dataset within a collection.
704 collections, optional.
705 An expression that fully or partially identifies the collections to
706 search for the dataset; see
707 :ref:`daf_butler_collection_expressions` for more information.
708 Defaults to ``self.defaults.collections``.
709 timespan : `Timespan`, optional
710 A timespan that the validity range of the dataset must overlap.
711 If not provided, any `~CollectionType.CALIBRATION` collections
712 matched by the ``collections`` argument will not be searched.
713 **kwargs
714 Additional keyword arguments passed to
715 `DataCoordinate.standardize` to convert ``dataId`` to a true
716 `DataCoordinate` or augment an existing one.
718 Returns
719 -------
720 ref : `DatasetRef`
721 A reference to the dataset, or `None` if no matching Dataset
722 was found.
724 Raises
725 ------
726 NoDefaultCollectionError
727 Raised if ``collections`` is `None` and
728 ``self.defaults.collections`` is `None`.
729 LookupError
730 Raised if one or more data ID keys are missing.
731 MissingDatasetTypeError
732 Raised if the dataset type does not exist.
733 MissingCollectionError
734 Raised if any of ``collections`` does not exist in the registry.
736 Notes
737 -----
738 This method simply returns `None` and does not raise an exception even
739 when the set of collections searched is intrinsically incompatible with
740 the dataset type, e.g. if ``datasetType.isCalibration() is False``, but
741 only `~CollectionType.CALIBRATION` collections are being searched.
742 This may make it harder to debug some lookup failures, but the behavior
743 is intentional; we consider it more important that failed searches are
744 reported consistently, regardless of the reason, and that adding
745 additional collections that do not contain a match to the search path
746 never changes the behavior.
748 This method handles component dataset types automatically, though most
749 other registry operations do not.
750 """
751 raise NotImplementedError()
753 @abstractmethod
754 def insertDatasets(
755 self,
756 datasetType: Union[DatasetType, str],
757 dataIds: Iterable[DataId],
758 run: Optional[str] = None,
759 expand: bool = True,
760 idGenerationMode: DatasetIdGenEnum = DatasetIdGenEnum.UNIQUE,
761 ) -> List[DatasetRef]:
762 """Insert one or more datasets into the `Registry`
764 This always adds new datasets; to associate existing datasets with
765 a new collection, use ``associate``.
767 Parameters
768 ----------
769 datasetType : `DatasetType` or `str`
770 A `DatasetType` or the name of one.
771 dataIds : `~collections.abc.Iterable` of `dict` or `DataCoordinate`
772 Dimension-based identifiers for the new datasets.
773 run : `str`, optional
774 The name of the run that produced the datasets. Defaults to
775 ``self.defaults.run``.
776 expand : `bool`, optional
777 If `True` (default), expand data IDs as they are inserted. This is
778 necessary in general to allow datastore to generate file templates,
779 but it may be disabled if the caller can guarantee this is
780 unnecessary.
781 idGenerationMode : `DatasetIdGenEnum`, optional
782 Specifies option for generating dataset IDs. By default unique IDs
783 are generated for each inserted dataset.
785 Returns
786 -------
787 refs : `list` of `DatasetRef`
788 Resolved `DatasetRef` instances for all given data IDs (in the same
789 order).
791 Raises
792 ------
793 DatasetTypeError
794 Raised if ``datasetType`` is not known to registry.
795 CollectionTypeError
796 Raised if ``run`` collection type is not `~CollectionType.RUN`.
797 NoDefaultCollectionError
798 Raised if ``run`` is `None` and ``self.defaults.run`` is `None`.
799 ConflictingDefinitionError
800 If a dataset with the same dataset type and data ID as one of those
801 given already exists in ``run``.
802 MissingCollectionError
803 Raised if ``run`` does not exist in the registry.
804 """
805 raise NotImplementedError()
807 @abstractmethod
808 def _importDatasets(
809 self,
810 datasets: Iterable[DatasetRef],
811 expand: bool = True,
812 idGenerationMode: DatasetIdGenEnum = DatasetIdGenEnum.UNIQUE,
813 reuseIds: bool = False,
814 ) -> List[DatasetRef]:
815 """Import one or more datasets into the `Registry`.
817 Difference from `insertDatasets` method is that this method accepts
818 `DatasetRef` instances which should already be resolved and have a
819 dataset ID. If registry supports globally-unique dataset IDs (e.g.
820 `uuid.UUID`) then datasets which already exist in the registry will be
821 ignored if imported again.
823 Parameters
824 ----------
825 datasets : `~collections.abc.Iterable` of `DatasetRef`
826 Datasets to be inserted. All `DatasetRef` instances must have
827 identical ``datasetType`` and ``run`` attributes. ``run``
828 attribute can be `None` and defaults to ``self.defaults.run``.
829 Datasets can specify ``id`` attribute which will be used for
830 inserted datasets. All dataset IDs must have the same type
831 (`int` or `uuid.UUID`), if type of dataset IDs does not match
832 configured backend then IDs will be ignored and new IDs will be
833 generated by backend.
834 expand : `bool`, optional
835 If `True` (default), expand data IDs as they are inserted. This is
836 necessary in general to allow datastore to generate file templates,
837 but it may be disabled if the caller can guarantee this is
838 unnecessary.
839 idGenerationMode : `DatasetIdGenEnum`, optional
840 Specifies option for generating dataset IDs when IDs are not
841 provided or their type does not match backend type. By default
842 unique IDs are generated for each inserted dataset.
843 reuseIds : `bool`, optional
844 If `True` then forces re-use of imported dataset IDs for integer
845 IDs which are normally generated as auto-incremented; exception
846 will be raised if imported IDs clash with existing ones. This
847 option has no effect on the use of globally-unique IDs which are
848 always re-used (or generated if integer IDs are being imported).
850 Returns
851 -------
852 refs : `list` of `DatasetRef`
853 Resolved `DatasetRef` instances for all given data IDs (in the same
854 order). If any of ``datasets`` has an ID which already exists in
855 the database then it will not be inserted or updated, but a
856 resolved `DatasetRef` will be returned for it in any case.
858 Raises
859 ------
860 NoDefaultCollectionError
861 Raised if ``run`` is `None` and ``self.defaults.run`` is `None`.
862 DatasetTypeError
863 Raised if datasets correspond to more than one dataset type or
864 dataset type is not known to registry.
865 ConflictingDefinitionError
866 If a dataset with the same dataset type and data ID as one of those
867 given already exists in ``run``.
868 MissingCollectionError
869 Raised if ``run`` does not exist in the registry.
871 Notes
872 -----
873 This method is considered package-private and internal to Butler
874 implementation. Clients outside daf_butler package should not use this
875 method.
876 """
877 raise NotImplementedError()
879 @abstractmethod
880 def getDataset(self, id: DatasetId) -> Optional[DatasetRef]:
881 """Retrieve a Dataset entry.
883 Parameters
884 ----------
885 id : `DatasetId`
886 The unique identifier for the dataset.
888 Returns
889 -------
890 ref : `DatasetRef` or `None`
891 A ref to the Dataset, or `None` if no matching Dataset
892 was found.
893 """
894 raise NotImplementedError()
896 @abstractmethod
897 def removeDatasets(self, refs: Iterable[DatasetRef]) -> None:
898 """Remove datasets from the Registry.
900 The datasets will be removed unconditionally from all collections, and
901 any `Quantum` that consumed this dataset will instead be marked with
902 having a NULL input. `Datastore` records will *not* be deleted; the
903 caller is responsible for ensuring that the dataset has already been
904 removed from all Datastores.
906 Parameters
907 ----------
908 refs : `Iterable` of `DatasetRef`
909 References to the datasets to be removed. Must include a valid
910 ``id`` attribute, and should be considered invalidated upon return.
912 Raises
913 ------
914 AmbiguousDatasetError
915 Raised if any ``ref.id`` is `None`.
916 OrphanedRecordError
917 Raised if any dataset is still present in any `Datastore`.
918 """
919 raise NotImplementedError()
921 @abstractmethod
922 def associate(self, collection: str, refs: Iterable[DatasetRef]) -> None:
923 """Add existing datasets to a `~CollectionType.TAGGED` collection.
925 If a DatasetRef with the same exact ID is already in a collection
926 nothing is changed. If a `DatasetRef` with the same `DatasetType` and
927 data ID but with different ID exists in the collection,
928 `ConflictingDefinitionError` is raised.
930 Parameters
931 ----------
932 collection : `str`
933 Indicates the collection the datasets should be associated with.
934 refs : `Iterable` [ `DatasetRef` ]
935 An iterable of resolved `DatasetRef` instances that already exist
936 in this `Registry`.
938 Raises
939 ------
940 ConflictingDefinitionError
941 If a Dataset with the given `DatasetRef` already exists in the
942 given collection.
943 AmbiguousDatasetError
944 Raised if ``any(ref.id is None for ref in refs)``.
945 MissingCollectionError
946 Raised if ``collection`` does not exist in the registry.
947 CollectionTypeError
948 Raise adding new datasets to the given ``collection`` is not
949 allowed.
950 """
951 raise NotImplementedError()
953 @abstractmethod
954 def disassociate(self, collection: str, refs: Iterable[DatasetRef]) -> None:
955 """Remove existing datasets from a `~CollectionType.TAGGED` collection.
957 ``collection`` and ``ref`` combinations that are not currently
958 associated are silently ignored.
960 Parameters
961 ----------
962 collection : `str`
963 The collection the datasets should no longer be associated with.
964 refs : `Iterable` [ `DatasetRef` ]
965 An iterable of resolved `DatasetRef` instances that already exist
966 in this `Registry`.
968 Raises
969 ------
970 AmbiguousDatasetError
971 Raised if any of the given dataset references is unresolved.
972 MissingCollectionError
973 Raised if ``collection`` does not exist in the registry.
974 CollectionTypeError
975 Raise adding new datasets to the given ``collection`` is not
976 allowed.
977 """
978 raise NotImplementedError()
980 @abstractmethod
981 def certify(self, collection: str, refs: Iterable[DatasetRef], timespan: Timespan) -> None:
982 """Associate one or more datasets with a calibration collection and a
983 validity range within it.
985 Parameters
986 ----------
987 collection : `str`
988 The name of an already-registered `~CollectionType.CALIBRATION`
989 collection.
990 refs : `Iterable` [ `DatasetRef` ]
991 Datasets to be associated.
992 timespan : `Timespan`
993 The validity range for these datasets within the collection.
995 Raises
996 ------
997 AmbiguousDatasetError
998 Raised if any of the given `DatasetRef` instances is unresolved.
999 ConflictingDefinitionError
1000 Raised if the collection already contains a different dataset with
1001 the same `DatasetType` and data ID and an overlapping validity
1002 range.
1003 CollectionTypeError
1004 Raised if ``collection`` is not a `~CollectionType.CALIBRATION`
1005 collection or if one or more datasets are of a dataset type for
1006 which `DatasetType.isCalibration` returns `False`.
1007 """
1008 raise NotImplementedError()
1010 @abstractmethod
1011 def decertify(
1012 self,
1013 collection: str,
1014 datasetType: Union[str, DatasetType],
1015 timespan: Timespan,
1016 *,
1017 dataIds: Optional[Iterable[DataId]] = None,
1018 ) -> None:
1019 """Remove or adjust datasets to clear a validity range within a
1020 calibration collection.
1022 Parameters
1023 ----------
1024 collection : `str`
1025 The name of an already-registered `~CollectionType.CALIBRATION`
1026 collection.
1027 datasetType : `str` or `DatasetType`
1028 Name or `DatasetType` instance for the datasets to be decertified.
1029 timespan : `Timespan`, optional
1030 The validity range to remove datasets from within the collection.
1031 Datasets that overlap this range but are not contained by it will
1032 have their validity ranges adjusted to not overlap it, which may
1033 split a single dataset validity range into two.
1034 dataIds : `Iterable` [ `DataId` ], optional
1035 Data IDs that should be decertified within the given validity range
1036 If `None`, all data IDs for ``self.datasetType`` will be
1037 decertified.
1039 Raises
1040 ------
1041 CollectionTypeError
1042 Raised if ``collection`` is not a `~CollectionType.CALIBRATION`
1043 collection or if ``datasetType.isCalibration() is False``.
1044 """
1045 raise NotImplementedError()
1047 @abstractmethod
1048 def getDatastoreBridgeManager(self) -> DatastoreRegistryBridgeManager:
1049 """Return an object that allows a new `Datastore` instance to
1050 communicate with this `Registry`.
1052 Returns
1053 -------
1054 manager : `DatastoreRegistryBridgeManager`
1055 Object that mediates communication between this `Registry` and its
1056 associated datastores.
1057 """
1058 raise NotImplementedError()
1060 @abstractmethod
1061 def getDatasetLocations(self, ref: DatasetRef) -> Iterable[str]:
1062 """Retrieve datastore locations for a given dataset.
1064 Parameters
1065 ----------
1066 ref : `DatasetRef`
1067 A reference to the dataset for which to retrieve storage
1068 information.
1070 Returns
1071 -------
1072 datastores : `Iterable` [ `str` ]
1073 All the matching datastores holding this dataset.
1075 Raises
1076 ------
1077 AmbiguousDatasetError
1078 Raised if ``ref.id`` is `None`.
1079 """
1080 raise NotImplementedError()
1082 @abstractmethod
1083 def expandDataId(
1084 self,
1085 dataId: Optional[DataId] = None,
1086 *,
1087 graph: Optional[DimensionGraph] = None,
1088 records: Optional[NameLookupMapping[DimensionElement, Optional[DimensionRecord]]] = None,
1089 withDefaults: bool = True,
1090 **kwargs: Any,
1091 ) -> DataCoordinate:
1092 """Expand a dimension-based data ID to include additional information.
1094 Parameters
1095 ----------
1096 dataId : `DataCoordinate` or `dict`, optional
1097 Data ID to be expanded; augmented and overridden by ``kwargs``.
1098 graph : `DimensionGraph`, optional
1099 Set of dimensions for the expanded ID. If `None`, the dimensions
1100 will be inferred from the keys of ``dataId`` and ``kwargs``.
1101 Dimensions that are in ``dataId`` or ``kwargs`` but not in
1102 ``graph`` are silently ignored, providing a way to extract and
1103 ``graph`` expand a subset of a data ID.
1104 records : `Mapping` [`str`, `DimensionRecord`], optional
1105 Dimension record data to use before querying the database for that
1106 data, keyed by element name.
1107 withDefaults : `bool`, optional
1108 Utilize ``self.defaults.dataId`` to fill in missing governor
1109 dimension key-value pairs. Defaults to `True` (i.e. defaults are
1110 used).
1111 **kwargs
1112 Additional keywords are treated like additional key-value pairs for
1113 ``dataId``, extending and overriding
1115 Returns
1116 -------
1117 expanded : `DataCoordinate`
1118 A data ID that includes full metadata for all of the dimensions it
1119 identifies, i.e. guarantees that ``expanded.hasRecords()`` and
1120 ``expanded.hasFull()`` both return `True`.
1122 Raises
1123 ------
1124 DataIdError
1125 Raised when ``dataId`` or keyword arguments specify unknown
1126 dimensions or values, or when a resulting data ID contains
1127 contradictory key-value pairs, according to dimension
1128 relationships.
1130 Notes
1131 -----
1132 This method cannot be relied upon to reject invalid data ID values
1133 for dimensions that do actually not have any record columns. For
1134 efficiency reasons the records for these dimensions (which have only
1135 dimension key values that are given by the caller) may be constructed
1136 directly rather than obtained from the registry database.
1137 """
1138 raise NotImplementedError()
1140 @abstractmethod
1141 def insertDimensionData(
1142 self,
1143 element: Union[DimensionElement, str],
1144 *data: Union[Mapping[str, Any], DimensionRecord],
1145 conform: bool = True,
1146 replace: bool = False,
1147 skip_existing: bool = False,
1148 ) -> None:
1149 """Insert one or more dimension records into the database.
1151 Parameters
1152 ----------
1153 element : `DimensionElement` or `str`
1154 The `DimensionElement` or name thereof that identifies the table
1155 records will be inserted into.
1156 data : `dict` or `DimensionRecord` (variadic)
1157 One or more records to insert.
1158 conform : `bool`, optional
1159 If `False` (`True` is default) perform no checking or conversions,
1160 and assume that ``element`` is a `DimensionElement` instance and
1161 ``data`` is a one or more `DimensionRecord` instances of the
1162 appropriate subclass.
1163 replace : `bool`, optional
1164 If `True` (`False` is default), replace existing records in the
1165 database if there is a conflict.
1166 skip_existing : `bool`, optional
1167 If `True` (`False` is default), skip insertion if a record with
1168 the same primary key values already exists. Unlike
1169 `syncDimensionData`, this will not detect when the given record
1170 differs from what is in the database, and should not be used when
1171 this is a concern.
1172 """
1173 raise NotImplementedError()
1175 @abstractmethod
1176 def syncDimensionData(
1177 self,
1178 element: Union[DimensionElement, str],
1179 row: Union[Mapping[str, Any], DimensionRecord],
1180 conform: bool = True,
1181 update: bool = False,
1182 ) -> Union[bool, Dict[str, Any]]:
1183 """Synchronize the given dimension record with the database, inserting
1184 if it does not already exist and comparing values if it does.
1186 Parameters
1187 ----------
1188 element : `DimensionElement` or `str`
1189 The `DimensionElement` or name thereof that identifies the table
1190 records will be inserted into.
1191 row : `dict` or `DimensionRecord`
1192 The record to insert.
1193 conform : `bool`, optional
1194 If `False` (`True` is default) perform no checking or conversions,
1195 and assume that ``element`` is a `DimensionElement` instance and
1196 ``data`` is a one or more `DimensionRecord` instances of the
1197 appropriate subclass.
1198 update: `bool`, optional
1199 If `True` (`False` is default), update the existing record in the
1200 database if there is a conflict.
1202 Returns
1203 -------
1204 inserted_or_updated : `bool` or `dict`
1205 `True` if a new row was inserted, `False` if no changes were
1206 needed, or a `dict` mapping updated column names to their old
1207 values if an update was performed (only possible if
1208 ``update=True``).
1210 Raises
1211 ------
1212 ConflictingDefinitionError
1213 Raised if the record exists in the database (according to primary
1214 key lookup) but is inconsistent with the given one.
1215 """
1216 raise NotImplementedError()
1218 @abstractmethod
1219 def queryDatasetTypes(
1220 self,
1221 expression: Any = ...,
1222 *,
1223 components: Optional[bool] = None,
1224 missing: Optional[List[str]] = None,
1225 ) -> Iterable[DatasetType]:
1226 """Iterate over the dataset types whose names match an expression.
1228 Parameters
1229 ----------
1230 expression : `Any`, optional
1231 An expression that fully or partially identifies the dataset types
1232 to return, such as a `str`, `re.Pattern`, or iterable thereof.
1233 ``...`` can be used to return all dataset types, and is the
1234 default. See :ref:`daf_butler_dataset_type_expressions` for more
1235 information.
1236 components : `bool`, optional
1237 If `True`, apply all expression patterns to component dataset type
1238 names as well. If `False`, never apply patterns to components.
1239 If `None` (default), apply patterns to components only if their
1240 parent datasets were not matched by the expression.
1241 Fully-specified component datasets (`str` or `DatasetType`
1242 instances) are always included.
1244 Values other than `False` are deprecated, and only `False` will be
1245 supported after v26. After v27 this argument will be removed
1246 entirely.
1247 missing : `list` of `str`, optional
1248 String dataset type names that were explicitly given (i.e. not
1249 regular expression patterns) but not found will be appended to this
1250 list, if it is provided.
1252 Returns
1253 -------
1254 dataset_types : `Iterable` [ `DatasetType`]
1255 An `Iterable` of `DatasetType` instances whose names match
1256 ``expression``.
1258 Raises
1259 ------
1260 DatasetTypeExpressionError
1261 Raised when ``expression`` is invalid.
1262 """
1263 raise NotImplementedError()
1265 @abstractmethod
1266 def queryCollections(
1267 self,
1268 expression: Any = ...,
1269 datasetType: Optional[DatasetType] = None,
1270 collectionTypes: Union[Iterable[CollectionType], CollectionType] = CollectionType.all(),
1271 flattenChains: bool = False,
1272 includeChains: Optional[bool] = None,
1273 ) -> Sequence[str]:
1274 """Iterate over the collections whose names match an expression.
1276 Parameters
1277 ----------
1278 expression : `Any`, optional
1279 An expression that identifies the collections to return, such as
1280 a `str` (for full matches or partial matches via globs),
1281 `re.Pattern` (for partial matches), or iterable thereof. ``...``
1282 can be used to return all collections, and is the default.
1283 See :ref:`daf_butler_collection_expressions` for more information.
1284 datasetType : `DatasetType`, optional
1285 If provided, only yield collections that may contain datasets of
1286 this type. This is a conservative approximation in general; it may
1287 yield collections that do not have any such datasets.
1288 collectionTypes : `AbstractSet` [ `CollectionType` ] or \
1289 `CollectionType`, optional
1290 If provided, only yield collections of these types.
1291 flattenChains : `bool`, optional
1292 If `True` (`False` is default), recursively yield the child
1293 collections of matching `~CollectionType.CHAINED` collections.
1294 includeChains : `bool`, optional
1295 If `True`, yield records for matching `~CollectionType.CHAINED`
1296 collections. Default is the opposite of ``flattenChains``: include
1297 either CHAINED collections or their children, but not both.
1299 Returns
1300 -------
1301 collections : `Sequence` [ `str` ]
1302 The names of collections that match ``expression``.
1304 Raises
1305 ------
1306 CollectionExpressionError
1307 Raised when ``expression`` is invalid.
1309 Notes
1310 -----
1311 The order in which collections are returned is unspecified, except that
1312 the children of a `~CollectionType.CHAINED` collection are guaranteed
1313 to be in the order in which they are searched. When multiple parent
1314 `~CollectionType.CHAINED` collections match the same criteria, the
1315 order in which the two lists appear is unspecified, and the lists of
1316 children may be incomplete if a child has multiple parents.
1317 """
1318 raise NotImplementedError()
1320 @abstractmethod
1321 def queryDatasets(
1322 self,
1323 datasetType: Any,
1324 *,
1325 collections: CollectionArgType | None = None,
1326 dimensions: Optional[Iterable[Union[Dimension, str]]] = None,
1327 dataId: Optional[DataId] = None,
1328 where: str = "",
1329 findFirst: bool = False,
1330 components: Optional[bool] = None,
1331 bind: Optional[Mapping[str, Any]] = None,
1332 check: bool = True,
1333 **kwargs: Any,
1334 ) -> DatasetQueryResults:
1335 """Query for and iterate over dataset references matching user-provided
1336 criteria.
1338 Parameters
1339 ----------
1340 datasetType
1341 An expression that fully or partially identifies the dataset types
1342 to be queried. Allowed types include `DatasetType`, `str`,
1343 `re.Pattern`, and iterables thereof. The special value ``...`` can
1344 be used to query all dataset types. See
1345 :ref:`daf_butler_dataset_type_expressions` for more information.
1346 collections: optional
1347 An expression that identifies the collections to search, such as a
1348 `str` (for full matches or partial matches via globs), `re.Pattern`
1349 (for partial matches), or iterable thereof. ``...`` can be used to
1350 search all collections (actually just all `~CollectionType.RUN`
1351 collections, because this will still find all datasets).
1352 If not provided, ``self.default.collections`` is used. See
1353 :ref:`daf_butler_collection_expressions` for more information.
1354 dimensions : `~collections.abc.Iterable` of `Dimension` or `str`
1355 Dimensions to include in the query (in addition to those used
1356 to identify the queried dataset type(s)), either to constrain
1357 the resulting datasets to those for which a matching dimension
1358 exists, or to relate the dataset type's dimensions to dimensions
1359 referenced by the ``dataId`` or ``where`` arguments.
1360 dataId : `dict` or `DataCoordinate`, optional
1361 A data ID whose key-value pairs are used as equality constraints
1362 in the query.
1363 where : `str`, optional
1364 A string expression similar to a SQL WHERE clause. May involve
1365 any column of a dimension table or (as a shortcut for the primary
1366 key column of a dimension table) dimension name. See
1367 :ref:`daf_butler_dimension_expressions` for more information.
1368 findFirst : `bool`, optional
1369 If `True` (`False` is default), for each result data ID, only
1370 yield one `DatasetRef` of each `DatasetType`, from the first
1371 collection in which a dataset of that dataset type appears
1372 (according to the order of ``collections`` passed in). If `True`,
1373 ``collections`` must not contain regular expressions and may not
1374 be ``...``.
1375 components : `bool`, optional
1376 If `True`, apply all dataset expression patterns to component
1377 dataset type names as well. If `False`, never apply patterns to
1378 components. If `None` (default), apply patterns to components only
1379 if their parent datasets were not matched by the expression.
1380 Fully-specified component datasets (`str` or `DatasetType`
1381 instances) are always included.
1383 Values other than `False` are deprecated, and only `False` will be
1384 supported after v26. After v27 this argument will be removed
1385 entirely.
1386 bind : `Mapping`, optional
1387 Mapping containing literal values that should be injected into the
1388 ``where`` expression, keyed by the identifiers they replace.
1389 check : `bool`, optional
1390 If `True` (default) check the query for consistency before
1391 executing it. This may reject some valid queries that resemble
1392 common mistakes (e.g. queries for visits without specifying an
1393 instrument).
1394 **kwargs
1395 Additional keyword arguments are forwarded to
1396 `DataCoordinate.standardize` when processing the ``dataId``
1397 argument (and may be used to provide a constraining data ID even
1398 when the ``dataId`` argument is `None`).
1400 Returns
1401 -------
1402 refs : `queries.DatasetQueryResults`
1403 Dataset references matching the given query criteria. Nested data
1404 IDs are guaranteed to include values for all implied dimensions
1405 (i.e. `DataCoordinate.hasFull` will return `True`), but will not
1406 include dimension records (`DataCoordinate.hasRecords` will be
1407 `False`) unless `~queries.DatasetQueryResults.expanded` is called
1408 on the result object (which returns a new one).
1410 Raises
1411 ------
1412 DatasetTypeExpressionError
1413 Raised when ``datasetType`` expression is invalid.
1414 TypeError
1415 Raised when the arguments are incompatible, such as when a
1416 collection wildcard is passed when ``findFirst`` is `True`, or
1417 when ``collections`` is `None` and``self.defaults.collections`` is
1418 also `None`.
1419 DataIdError
1420 Raised when ``dataId`` or keyword arguments specify unknown
1421 dimensions or values, or when they contain inconsistent values.
1422 UserExpressionError
1423 Raised when ``where`` expression is invalid.
1425 Notes
1426 -----
1427 When multiple dataset types are queried in a single call, the
1428 results of this operation are equivalent to querying for each dataset
1429 type separately in turn, and no information about the relationships
1430 between datasets of different types is included. In contexts where
1431 that kind of information is important, the recommended pattern is to
1432 use `queryDataIds` to first obtain data IDs (possibly with the
1433 desired dataset types and collections passed as constraints to the
1434 query), and then use multiple (generally much simpler) calls to
1435 `queryDatasets` with the returned data IDs passed as constraints.
1436 """
1437 raise NotImplementedError()
1439 @abstractmethod
1440 def queryDataIds(
1441 self,
1442 dimensions: Union[Iterable[Union[Dimension, str]], Dimension, str],
1443 *,
1444 dataId: Optional[DataId] = None,
1445 datasets: Any = None,
1446 collections: CollectionArgType | None = None,
1447 where: str = "",
1448 components: Optional[bool] = None,
1449 bind: Optional[Mapping[str, Any]] = None,
1450 check: bool = True,
1451 **kwargs: Any,
1452 ) -> DataCoordinateQueryResults:
1453 """Query for data IDs matching user-provided criteria.
1455 Parameters
1456 ----------
1457 dimensions : `Dimension` or `str`, or iterable thereof
1458 The dimensions of the data IDs to yield, as either `Dimension`
1459 instances or `str`. Will be automatically expanded to a complete
1460 `DimensionGraph`.
1461 dataId : `dict` or `DataCoordinate`, optional
1462 A data ID whose key-value pairs are used as equality constraints
1463 in the query.
1464 datasets : `Any`, optional
1465 An expression that fully or partially identifies dataset types
1466 that should constrain the yielded data IDs. For example, including
1467 "raw" here would constrain the yielded ``instrument``,
1468 ``exposure``, ``detector``, and ``physical_filter`` values to only
1469 those for which at least one "raw" dataset exists in
1470 ``collections``. Allowed types include `DatasetType`, `str`,
1471 and iterables thereof. Regular expression objects (i.e.
1472 `re.Pattern`) are deprecated and will be removed after the v26
1473 release. See :ref:`daf_butler_dataset_type_expressions` for more
1474 information.
1475 collections: `Any`, optional
1476 An expression that identifies the collections to search for
1477 datasets, such as a `str` (for full matches or partial matches
1478 via globs), `re.Pattern` (for partial matches), or iterable
1479 thereof. ``...`` can be used to search all collections (actually
1480 just all `~CollectionType.RUN` collections, because this will
1481 still find all datasets). If not provided,
1482 ``self.default.collections`` is used. Ignored unless ``datasets``
1483 is also passed. See :ref:`daf_butler_collection_expressions` for
1484 more information.
1485 where : `str`, optional
1486 A string expression similar to a SQL WHERE clause. May involve
1487 any column of a dimension table or (as a shortcut for the primary
1488 key column of a dimension table) dimension name. See
1489 :ref:`daf_butler_dimension_expressions` for more information.
1490 components : `bool`, optional
1491 If `True`, apply all dataset expression patterns to component
1492 dataset type names as well. If `False`, never apply patterns to
1493 components. If `None` (default), apply patterns to components only
1494 if their parent datasets were not matched by the expression.
1495 Fully-specified component datasets (`str` or `DatasetType`
1496 instances) are always included.
1498 Values other than `False` are deprecated, and only `False` will be
1499 supported after v26. After v27 this argument will be removed
1500 entirely.
1501 bind : `Mapping`, optional
1502 Mapping containing literal values that should be injected into the
1503 ``where`` expression, keyed by the identifiers they replace.
1504 check : `bool`, optional
1505 If `True` (default) check the query for consistency before
1506 executing it. This may reject some valid queries that resemble
1507 common mistakes (e.g. queries for visits without specifying an
1508 instrument).
1509 **kwargs
1510 Additional keyword arguments are forwarded to
1511 `DataCoordinate.standardize` when processing the ``dataId``
1512 argument (and may be used to provide a constraining data ID even
1513 when the ``dataId`` argument is `None`).
1515 Returns
1516 -------
1517 dataIds : `queries.DataCoordinateQueryResults`
1518 Data IDs matching the given query parameters. These are guaranteed
1519 to identify all dimensions (`DataCoordinate.hasFull` returns
1520 `True`), but will not contain `DimensionRecord` objects
1521 (`DataCoordinate.hasRecords` returns `False`). Call
1522 `DataCoordinateQueryResults.expanded` on the returned object to
1523 fetch those (and consider using
1524 `DataCoordinateQueryResults.materialize` on the returned object
1525 first if the expected number of rows is very large). See
1526 documentation for those methods for additional information.
1528 Raises
1529 ------
1530 NoDefaultCollectionError
1531 Raised if ``collections`` is `None` and
1532 ``self.defaults.collections`` is `None`.
1533 CollectionExpressionError
1534 Raised when ``collections`` expression is invalid.
1535 DataIdError
1536 Raised when ``dataId`` or keyword arguments specify unknown
1537 dimensions or values, or when they contain inconsistent values.
1538 DatasetTypeExpressionError
1539 Raised when ``datasetType`` expression is invalid.
1540 UserExpressionError
1541 Raised when ``where`` expression is invalid.
1542 """
1543 raise NotImplementedError()
1545 @abstractmethod
1546 def queryDimensionRecords(
1547 self,
1548 element: Union[DimensionElement, str],
1549 *,
1550 dataId: Optional[DataId] = None,
1551 datasets: Any = None,
1552 collections: CollectionArgType | None = None,
1553 where: str = "",
1554 components: Optional[bool] = None,
1555 bind: Optional[Mapping[str, Any]] = None,
1556 check: bool = True,
1557 **kwargs: Any,
1558 ) -> DimensionRecordQueryResults:
1559 """Query for dimension information matching user-provided criteria.
1561 Parameters
1562 ----------
1563 element : `DimensionElement` or `str`
1564 The dimension element to obtain records for.
1565 dataId : `dict` or `DataCoordinate`, optional
1566 A data ID whose key-value pairs are used as equality constraints
1567 in the query.
1568 datasets : `Any`, optional
1569 An expression that fully or partially identifies dataset types
1570 that should constrain the yielded records. See `queryDataIds` and
1571 :ref:`daf_butler_dataset_type_expressions` for more information.
1572 collections : `Any`, optional
1573 An expression that identifies the collections to search for
1574 datasets, such as a `str` (for full matches or partial matches
1575 via globs), `re.Pattern` (for partial matches), or iterable
1576 thereof. ``...`` can be used to search all collections (actually
1577 just all `~CollectionType.RUN` collections, because this will
1578 still find all datasets). If not provided,
1579 ``self.default.collections`` is used. Ignored unless ``datasets``
1580 is also passed. See :ref:`daf_butler_collection_expressions` for
1581 more information.
1582 where : `str`, optional
1583 A string expression similar to a SQL WHERE clause. See
1584 `queryDataIds` and :ref:`daf_butler_dimension_expressions` for more
1585 information.
1586 components : `bool`, optional
1587 Whether to apply dataset expressions to components as well.
1588 See `queryDataIds` for more information.
1590 Values other than `False` are deprecated, and only `False` will be
1591 supported after v26. After v27 this argument will be removed
1592 entirely.
1593 bind : `Mapping`, optional
1594 Mapping containing literal values that should be injected into the
1595 ``where`` expression, keyed by the identifiers they replace.
1596 check : `bool`, optional
1597 If `True` (default) check the query for consistency before
1598 executing it. This may reject some valid queries that resemble
1599 common mistakes (e.g. queries for visits without specifying an
1600 instrument).
1601 **kwargs
1602 Additional keyword arguments are forwarded to
1603 `DataCoordinate.standardize` when processing the ``dataId``
1604 argument (and may be used to provide a constraining data ID even
1605 when the ``dataId`` argument is `None`).
1607 Returns
1608 -------
1609 dataIds : `queries.DimensionRecordQueryResults`
1610 Data IDs matching the given query parameters.
1612 Raises
1613 ------
1614 NoDefaultCollectionError
1615 Raised if ``collections`` is `None` and
1616 ``self.defaults.collections`` is `None`.
1617 CollectionExpressionError
1618 Raised when ``collections`` expression is invalid.
1619 DataIdError
1620 Raised when ``dataId`` or keyword arguments specify unknown
1621 dimensions or values, or when they contain inconsistent values.
1622 DatasetTypeExpressionError
1623 Raised when ``datasetType`` expression is invalid.
1624 UserExpressionError
1625 Raised when ``where`` expression is invalid.
1626 """
1627 raise NotImplementedError()
1629 @abstractmethod
1630 def queryDatasetAssociations(
1631 self,
1632 datasetType: Union[str, DatasetType],
1633 collections: CollectionArgType | None = Ellipsis,
1634 *,
1635 collectionTypes: Iterable[CollectionType] = CollectionType.all(),
1636 flattenChains: bool = False,
1637 ) -> Iterator[DatasetAssociation]:
1638 """Iterate over dataset-collection combinations where the dataset is in
1639 the collection.
1641 This method is a temporary placeholder for better support for
1642 association results in `queryDatasets`. It will probably be
1643 removed in the future, and should be avoided in production code
1644 whenever possible.
1646 Parameters
1647 ----------
1648 datasetType : `DatasetType` or `str`
1649 A dataset type object or the name of one.
1650 collections: `Any`, optional
1651 An expression that identifies the collections to search for
1652 datasets, such as a `str` (for full matches or partial matches
1653 via globs), `re.Pattern` (for partial matches), or iterable
1654 thereof. ``...`` can be used to search all collections (actually
1655 just all `~CollectionType.RUN` collections, because this will still
1656 find all datasets). If not provided, ``self.default.collections``
1657 is used. See :ref:`daf_butler_collection_expressions` for more
1658 information.
1659 collectionTypes : `AbstractSet` [ `CollectionType` ], optional
1660 If provided, only yield associations from collections of these
1661 types.
1662 flattenChains : `bool`, optional
1663 If `True` (default) search in the children of
1664 `~CollectionType.CHAINED` collections. If `False`, ``CHAINED``
1665 collections are ignored.
1667 Yields
1668 ------
1669 association : `.DatasetAssociation`
1670 Object representing the relationship between a single dataset and
1671 a single collection.
1673 Raises
1674 ------
1675 NoDefaultCollectionError
1676 Raised if ``collections`` is `None` and
1677 ``self.defaults.collections`` is `None`.
1678 CollectionExpressionError
1679 Raised when ``collections`` expression is invalid.
1680 """
1681 raise NotImplementedError()
1683 @property
1684 def obsCoreTableManager(self) -> ObsCoreTableManager | None:
1685 """ObsCore manager instance for this registry (`ObsCoreTableManager`
1686 or `None`).
1688 ObsCore manager may not be implemented for all registry backend, or
1689 may not be enabled for many repositories.
1690 """
1691 return None
1693 storageClasses: StorageClassFactory
1694 """All storage classes known to the registry (`StorageClassFactory`).
1695 """
1697 datasetIdFactory: DatasetIdFactory
1698 """Factory for dataset IDs."""